Apple Adds Brain-to-Computer Protocol to Its Accessibility Repertoire By John P. Mello Jr. May 14, 2025 5:00 AM PT ADVERTISEMENT Rubrik Foward 2025: The Future of Cyber Resilience is here When an attacker comes for your business, will you be ready? Chart your path to cyber resilience and keep your business running. June 4 | Virtual Event | Register Now Among a raft of upcoming accessibility tools revealed Tuesday, Apple announced a new protocol for brain-to-computer interfaces (BCIs) within its Switch Control feature. The protocol allows iOS, iPadOS, and visionOS devices to support an emerging technology that enables users to control their digital hardware without physical movement. One of the first companies to take advantage of the new protocol will be New York-based Synchron. “This marks a major milestone in accessibility and neurotechnology, where users implanted with Synchron’s BCI can control iPhone, iPad, and Apple Vision Pro directly with their thoughts without the need for physical movement or voice commands,” the company said in a statement. It added that Synchron’s BCI system will seamlessly integrate with Apple’s built-in accessibility features, including Switch Control, giving users an intuitive way to use their devices and laying the foundation for a new generation of cognitive input technologies. “This marks a defining moment for human-device interaction,” Synchron CEO and Co-Founder Tom Oxley said in a statement. “BCI is more than an accessibility tool, it’s a next-generation interface layer.” “Apple is helping to pioneer a new interface paradigm, where brain signals are formally recognized alongside touch, voice, and typing,” he continued. “With BCI recognized as a native input for Apple devices, there are new possibilities for people living with paralysis and beyond.” BCI Validation Tetiana Aleksandrova, CEO of Subsense, a biotechnology company in Covina, Calif., specializing in non-surgical bidirectional brain-computer interfaces, maintained Apple’s announcement marks a powerful signal showing the evolution of BCI. “I see it as Apple throwing open the gates — a single-stroke move that invites clinically-validated BCIs like Synchron’s Stentrode to plug straight into a billion-device ecosystem,” she told TechNewsWorld. “For patients, it means ‘mind-to-message’ control without middleware,” she said. “For the BCI industry, it’s a public stamp that neural input is ready for prime time — and yes, that’s a thrilling milestone for all of us building the next generation of non-surgical systems. It’s a shift, moving BCI from a nascent technology to more mainstream applications.” Aleksandrova maintained that BCI fits nicely into Apple’s overall accessibility strategy. “Apple’s playbook is to solve an extreme edge case, polish the UX until it’s invisible, then let the rest of the world adopt it,” she explained. “VoiceOver paved the way for Siri. Switch Control turned into Face Gestures. BCI support is the natural next rung on that ladder. Accessibility isn’t a side quest for Apple — it’s the R and D lab that future-proofs their core UI.” “Apple devices put unlimited information at users’ fingertips,” she added, “but for people with disabilities from TBI or ALS, full access can be out of reach. BCI technology helps bridge that gap, giving them a way to control and interact with their devices using only their brain activity.” Analysts See BCI as Long-Term Technology Apple’s embrace of BCI is significant, but its impact still lies in the future, noted Will Kerwin, technology equity analyst with Morningstar Research Services in Chicago. “While a particularly cool announcement, we think this type of feature is a long way away from full commercialization and not material to investors in Apple at this point,” he told TechNewsWorld. Kerwin pointed out that Synchron’s Stentrode BCI currently only has a sample size of 10 people. “Long-term, yes, this technology could have huge implications for how humans interact with technology, and we see it as an adjacency to AI, where generative AI could help improve the interface and ability for humans to communicate via the implant,” he said. “But again, we see this as an extremely long-term journey in its nascent days.” According to the Wall Street Journal, FDA approval of Synchron’s Stenrode technology is years away. The procedure involves implanting a stent-mounted electrode array into a blood vessel in the brain, so there’s no need for open brain surgery. “Some companies in the BCI space are focused on cortical control of prosthetics, others on cognitive enhancement or memory restoration,” Synchron spokesperson Kimberly Ha told TechNewsWorld. “What sets us apart is our focus on scalability and safety. By implanting via the blood vessels, we avoid open brain surgery, making our approach more feasible for potentially broader medical use.” Solving the BCI Scalability Problem Ha acknowledged that there are significant challenges to the broad adoption of BCI. “Scalability is one of the biggest,” she said. “Historically, many BCI systems have required open brain surgery, which presents serious risks and limits to who can access the technology,” she explained. “It’s simply not scalable for widespread clinical or consumer use.” “Synchron takes a fundamentally different approach,” she continued. “Our Stentrode device is implanted via the blood vessels, similar to a heart stent, avoiding the need to open the skull or directly penetrate brain tissue. This makes the procedure far less invasive, more accessible to patients, and better suited to real-world clinical deployment.” There are also challenges to developing the BCI apps themselves. “The biggest challenge in developing BCI applications is the trade-off between signal quality and accessibility,” Aleksandrova explained. “While a directly implanted BCI offers strong brain signals, surgery is risky. With non-invasive systems, the resolution is poor.” Her company, Subsense, is trying to offer the best of both worlds through the use of nanoparticles, which can provide bidirectional communication by crossing the blood-brain barrier to interact with neurons and transmit signals. Thought-Driven Interfaces for New Use Cases Ha noted that in addition to medical applications, BCI could be used for hands-free device control across all types of digital platforms, neuroadaptive gaming, or immersive XR experiences. “BCI opens doors to applications in mental wellness, communication tools, and cognitive enhancement,” Aleksandrova added. “You’ll fire off texts, browse AR screens, or write code just by thinking,” she said. “You’ll slip seamlessly into a drone or handle a surgical robot as though it were your own hand, and nudge your smart home with a silent impulse that dims the lights when focus peaks.” “Entertainment will read the room inside your head — dialing a game’s difficulty or a film’s plot to match your mood — while always-on neural metrics warn of fatigue, migraines, or anxiety hours before you notice and even surface names or ideas when your memory stalls,” she predicted. “Your unique brainwave ‘fingerprint’ will replace passwords, and researchers are already sketching ways to preserve those patterns so our minds can outlast failing bodies.” “I’m genuinely proud of Synchron and Apple for opening this door,” she said. John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. More by John P. Mello Jr. view all More in Science

AI Is Rewriting the Rules of Brand Management By John P. Mello Jr. May 13, 2025 5:00 AM PT ADVERTISEMENT Build HubSpot Apps, Faster New developer products preview the future of app building on HubSpot, including deeper extensibility, flexible UI, modern prototyping tools, and more. Learn More. Although building trust through a carefully crafted brand message is still important, artificial intelligence may be undermining its traditional influence. “AI isn’t just helping businesses create content or automate tasks; it’s empowering individuals to become instant digital detectives,” Mike Allton, chief storyteller at Agorapulse, a social media management platform for businesses, wrote Monday on LinkedIn. What that means, he explained, is a company’s entire digital history — reviews, articles, social media sentiment, even employee feedback — is now more transparent and instantly “queryable” than ever before. “The carefully crafted brand message? It’s still important, but AI can now cross-reference it with raw, aggregated public data in seconds,” he noted. Edwin Miller, CEO of Marchex, a conversation intelligence platform maker headquartered in Seattle, explained that the rise of large language models and real-time data analytics has effectively turned a company’s full digital footprint into a searchable, easy-to-interpret, and evaluative source of truth. “We’re entering a world where a company’s entire identity, how it treats customers, how it responds to criticism, what employees really think, and how well it delivers on its promises, can be surfaced instantly by AI,” he told TechNewsWorld. “And not just by researchers or journalists, but by consumers, investors, and competitors.” “This means companies no longer control the brand narrative the way they used to,” he said. “The narrative is now co-authored by customers, employees, and digital observers, with AI acting as a kind of omnipresent interpreter. That changes the playing field for brand management entirely.” AI Shrinks Trust-Building to Milliseconds Mark N. Vena, president and principal analyst for SmartTech Research in Las Vegas, argued that brand management is a “huge deal” in the AI age. “Brand management is no longer just about campaigns — it’s about constantly monitoring and reacting to a living, breathing digital footprint,” he told TechNewsWorld. “Every customer interaction, review, or leaked internal memo can instantly shape public perception,” he said. “That means brand managers must be part storyteller, part crisis manager, and fully agile. The brand isn’t what you say it is — it’s what the internet says it is.” Allton noted that AI’s capability to “vet” or “audit” is a powerful reminder that, as AI is integrated into businesses, they must also consider how the external AI ecosystem perceives them. “It’s no longer enough to say you’re trustworthy; the data must reflect it because that data is now incredibly accessible and interpretable by AI,” he wrote. “Trust used to be built over years and could be lost in moments,” added Lizi Sprague, co-founder of Songue PR, a public relations agency in San Francisco. “Now, with AI, trust can be verified in milliseconds. Every interaction, review, and employee comment becomes part of your permanent trust score.” She told TechNewsWorld: “AI isn’t replacing reputation managers or comms people; it’s making them more crucial than ever. In an AI-driven world, reputation management evolves from damage control to proactive narrative architecture.” Proactive Transparency Brand managers will also need to be more proactive. They need to pay attention to how their brand is represented in the most popular AI tools. “Brands should be conducting searches that test the way their reputation is represented or conveyed in those tools, and they should be paying attention to the sources that are referenced by AI tools,” said Damian Rollison, director of market insights at SOCi, a marketing solutions company in San Diego. “If a company focuses a lot on local marketing, they should be paying attention to reviews of a business in Google, Yelp, or TripAdvisor — those kinds of sources — all of which are heavily cited by AI,” he told TechNewsWorld. “If they’re not paying attention to those reviews and taking action to respond when consumers offer feedback — apologizing if they had a bad experience, offering some kind of remedy, thanking customers when they give you positive feedback — then they have even more reason than ever to pay attention to those reviews and respond to them now.” Dev Nag, CEO and founder of QueryPal, a customer support chatbot based in San Francisco, explained that an AI-searchable landscape will create persistent accountability. “Every ethical lapse, broken promise, and controversial statement lives on in digital archives, ready to be surfaced by AI at any moment,” he told TechNewsWorld. “Companies can leverage this AI-scrutinized environment by embracing proactive transparency,” he said. “Organizations should use AI tools to continuously monitor customer sentiment across vast data streams, gaining early warning of reputation risks and identifying improvement areas before issues escalate into crises.” New Era of AI-Driven Accountability Nag recommends conducting regular AI reputation audits, doubling down on authenticity, pursuing strong media coverage in respected outlets, empowering employees as reputation ambassadors, implementing AI monitoring with rapid response protocols, and preparing for AI-driven crises, including misinformation attacks. Transparency without controls, though, can harm a brand. “Doing reputation management well requires a tight focus on the behavior of those who can affect the appearance of the related firm,” said Rob Enderle, president and principal analyst of the Enderle Group, an advisory services firm in Bend, Ore. “If more transparency is created without these controls and training in place, coupled with strong execution, monitoring, and a strong crisis team, the outcome is likely to be catastrophic,” he told TechNewsWorld. “AI is now part of the reputation equation,” added Matthew A. Gilbert, a marketing lecturer at Coastal Carolina University. “It monitors everything, from customer reviews to employee comments,” he told TechNewsWorld. “Brands should treat it as an early warning system and act before issues escalate.” AI in Branding Demands Action, Not Panic Allton argued that the rise of AI as a reputation manager isn’t a cause for alarm but a cause for action. However, it does make some demands on businesses. They include: Non-Negotiable Radical Authenticity If there are inconsistencies between what your brand promises and what the public data reflects, AI-powered searches will likely highlight them. Your operations must genuinely align with your messaging.“Authenticity is no longer a decision made by brands regarding which cards to reveal; instead, it has become an inevitable force driven by the public, as everything will eventually come to light,” said Reilly Newman, founder and brand strategist at Motif Brands, a brand transformation company, in Paso Robles, Calif. “Authenticity is not merely a new initiative for brands,” he told TechNewsWorld. “It is a necessity and an expected element of any company.” The “AI Response” Is Your New First Impression For many, the first true understanding of your business might come from an AI-generated summary, Allton noted. What story is the collective data telling about you?Kseniya Melnikova, a marketing strategist with Melk PR, a marketing agency in Sioux Falls, S.D., recalled a client who believed their low engagement was due to a lack of clear marketing materials. “Using AI to analyze their community feedback, we discovered the real issue was that customers misunderstood who they were,” she told TechNewsWorld. “They were perceived as a retailer when, in fact, they were an insurance fulfillment service. With this insight, we produced fewer — but clearer — materials that corrected the misunderstanding and improved customer outcomes.” Human Values Still Drive the Core Code While AIs process the data, the data itself reflects human experiences and actions, Allton explained. Building a trustworthy business rooted in solid ethical practices provides the best input for any AI assessment.Brand Basics Businesses that stick to fundamentals, though, shouldn’t have to worry about the new unofficial reputation manager. “Companies need to deliver great products and services and back them up with strong support,” asserted Greg Sterling, co-founder of Near Media, a market research firm in San Francisco. “Marketing is a separate thing, but their core business and the way they treat their customers need to be very solid and reliable,” he told TechNewsWorld. “Marketing and brand campaigns can then be built on top of that fundamental authenticity and ethical conduct, which will be reflected in AI results.” “I think people get very confused about what makes a successful business, and they’re focused on tips and tricks and marketing manipulation,” he said. “Great marketing is built on great products and services. Great brands are built by delivering great products and services, being consistent, and treating customers well. That’s the core proposition that everything else flows out of.” John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. More by John P. Mello Jr. view all More in Artificial Intelligence

Hold onto your popcorn, folks, because artificial intelligence (AI) is about to take our beloved entertainment for a joyride, and it probably won’t ask for directions. We’re talking about a future where AI isn’t just fetching your slippers (it might do that too; who knows?) but is actively involved in crafting stories that make us laugh, cry, or hide behind the sofa. From eBooks that read you back, to movies that might just star you (unintentionally or otherwise), the digital script is being rewritten. So, grab your (soon-to-be AI-recommended) snack of choice, and let’s peek at what the future of fun, powered by our increasingly clever silicon sidekicks, might look like. We’ll close with my product of the week, the new Violet (well, really Purple) Surface Pro from Microsoft. AI Personalization and Entertainment Preferences Right now, AI in entertainment is like that friend who thinks they know your taste because you watched one cat video, and now your feed is an endless feline festival. Platforms like Netflix and Spotify are doing their best with AI-powered recommendations, trying to guess what movie or song will tickle your fancy. But bless their digital hearts, it’s often a bit of a crapshoot. Future AI, however, aims to be your entertainment soulmate. It won’t just know you like sci-fi; it’ll understand your particular craving for “1950s-style alien invasion comedies but with surprisingly good character development for the pug.” It’ll be like having a tiny, all-knowing librarian or film critic living in your phone, unearthing obscure indie eBooks, or that one bizarre international TV show that speaks directly to your quirky soul. Finally, a way to find stories so niche you’ll wonder if the AI just wrote them for you on the spot. AI-Driven Adaptation in Books and Movies Get ready for stories less set in stone and more like Silly Putty. For eBooks, imagine a novel that senses you’re doom-scrolling instead of reading. With your say-so, the AI might subtly tweak the plot: maybe that boring council meeting gets a surprise dragon attack, or that side character you secretly adore suddenly gets a heroic monologue. It’s not about the AI ghostwriting a new ending because it didn’t like the original, but more like offering a director’s cut tailored to your brainwaves. TV shows and streamed movies will get the same treatment. Is the family movie night hitting a snag because little Timmy is terrified of the friendly ghost? The AI might gently dim the spooky sound effects or add a reassuring voiceover from a cartoon squirrel. Or if you, a connoisseur of crime dramas, are ten steps ahead of the detective, the AI could weave in a red herring so fiendishly clever you’ll tip your fedora. Such AI-powered storytelling will craft narratives so adaptive they’ll feel like they were made just for your Tuesday night. AI Personalization in Film and TV Experiences Cinema night is about to get a whole lot more personal. First up, AI could sprinkle in some local flavor. Imagine watching a superhero movie, and the obligatory coffee shop scene features a digital version of your town’s beloved, slightly sticky doughnut emporium. But why stop there? The real fun begins with you in the picture. With your permission (and a waiver, probably), your face could appear digitally as an extra in the crowd scene, or your name might be subtly dropped by a minor character ordering a latte. “Table for Smith, party of four? Your intergalactic overlord will see you now.” For the truly adventurous, picture this: the AI gauges the audience’s collective gasps, laughs, or even synchronized eye-rolls and nudges the plot. “Hmm, they didn’t buy that plot twist; deploy alternate ending B!” This is way beyond just yelling at the screen; it’s the screen (politely) listening back, an extension of how AI already tries to understand viewer preferences and tailor content. Every showing could be a fresh adventure! Smarter NPCs and Adaptive Game Design Video games are where AI is really flexing its muscles, and the future looks hilariously immersive. Human-Like NPCs: Forget those Non-Player Characters who repeat the same three lines of dialogue even after you’ve saved their village from a meteor. Future NPCs will have memories. They’ll remember you “accidentally” sold their prize-winning turnip. They’ll hold grudges. They’ll form opinions. They might even start a book club without you. We’re talking NPCs with their own sitcom-worthy lives. Hyper-Realistic Avatars: Your in-game avatar won’t just look like a slightly shinier version of you; it’ll emote like you. Imagine trying to bluff your way through a space poker game when your avatar keeps nervously sweating digital bullets. AI will drive character customization to uncanny levels. Dynamically Altered Content: Is that boss battle making you want to throw your controller through the screen? A benevolent AI might subtly suggest a new strategy through an “ancient cryptic clue” (that it just generated) or maybe even trip the boss on a conveniently placed banana peel. The goal is to make games so adaptive that they practically read your mind, keeping you hooked instead of ready to throw your controller.. Observer Influence: Even watching games could become interactive. Imagine your Twitch chat collectively voting to unleash a flock of slightly confused pigeons into your friend’s meticulously planned stealth mission. Harmless chaos, managed by AI! AI Collaboration and Audience Contributions If your brilliant, shouted-at-the-screen suggestion for how the detective could finally solve the case makes its way into the dynamically adapting TV show and boosts ratings, should you get a cut? It’s a thought! As AI lets audiences nudge stories or even contribute game-changing ideas, we might see new ways to reward those golden nuggets of accidental genius. Forget royalties; maybe you get a “Narrative Nudge” bonus or your name in the AI-generated “Special Thanks to that Shouter in Row G” credits. Navigating IP and copyright will be a hilarious legal tangle. Still, a future where your brilliant shower thoughts actually improve a blockbuster isn’t totally off the table, ushering in a truly collaborative (and potentially litigious) entertainment ecosystem. Transmedia Storytelling Now, let’s get really wild. Imagine your day starts with an eBook whose plot thickens based on your morning mood — as determined by how aggressively you hit the snooze button. At a key moment, the eBook suggests, “Want to see how this daring escape really goes down?” Bam! You’re playing a short, intense video game sequence on your phone. Later, that game’s outcome subtly alters the character dynamics in the TV show adaptation you watch that evening. Your personal “entertainment AI,” — let’s call it “Hal” for kicks; what could go wrong? — learns your tastes across everything. It then weaves a grand, personalized narrative that flows between your Kindle, smart TV, game console, and even that interactive movie you saw last week. Characters you love could show up in unexpected places, their storylines intertwining across platforms, all shaped by your choices and the AI’s increasingly insightful (or hilariously off-base) understanding of what makes you tick. This transmedia storytelling, supercharged by AI, would be less like watching a story and more like living in your own, ever-changing, AI-fueled epic. Wrapping Up: AI’s Entertainment Future So, is AI coming for our entertainment? You bet your algorithm it is! It promises a future where stories are more personal, more interactive, and possibly more delightfully bizarre than ever before. There will be ethical quandaries, creative debates, and probably a few AI-generated movie scripts that are so bad they’re good. One thing’s for sure: the future of entertainment looks less like a pre-recorded show and more like an improvisational comedy act where we all, willingly or not, get to be part of the cast. Now, if you’ll excuse me, my AI just recommended a documentary about the history of sporks. Wish me luck. The New Microsoft Surface Pro: Purple Reign Surface Pro 12-inch, Violet (Image Credit: Microsoft) My wife and I have a thing for purple. Not just any purple, but that deep, vibrant, makes-you-smile kind of purple. So, imagine our royal delight when Microsoft unveiled its latest Surface Pro – the brand new 12-inch Surface Pro Copilot+ PC — and lo and behold, among the standard Platinum and a rather fetching ‘Ocean’ blue-green, sits a glorious Violet option. After years of Surface devices playing it safe with silvers and blacks, this one feels like our Surface. A Brief History Lesson Let’s be honest: the Surface line was born largely because the iPad was eating everyone’s lunch back in 2012. Microsoft’s vision was ambitious: create a tablet that could truly replace your laptop, running full-fat Windows. It was a noble quest, but the path was rocky. Remember Windows RT? Shudder. For years, Surface devices felt like brilliant ideas slightly hampered by compromises: battery life, app compatibility (especially on early Arm versions), or trying to convince people they weren’t just chunkier iPads. Powerful and versatile, yes — but not always fun. Beyond a Color Refresh, It’s a True Upgrade This new 12-inch Surface Pro feels different. It’s incredibly thin and light, rocking a new Arm-based Snapdragon X Plus processor. Now, Arm on Windows has had a checkered past, but this is part of the new “Copilot+ PC” wave, promising significant performance leaps and killer battery life (up to 16 hours claimed!), plus integrated AI features like Recall that might actually be useful; or slightly creepy. The jury’s still out. Coupled with the excellent (and still sold separately, sigh) Surface Pro Keyboard, this finally feels like the ultra-portable, do-anything Windows machine we’ve been waiting for — and did I mention it comes in Violet? It just ties the whole elegant package together. Who Should Consider the New Surface Pro? Okay, besides people like us with impeccable taste in colors, who is this for? It’s ideal for mobile professionals who need the full power of Windows desktop apps in a super-portable form factor. Students who need something light for lectures but can run serious software will love it. Creatives who prefer Windows-based tools over iPadOS limitations should definitely take a look. If you want the ultimate blend of tablet portability and laptop capability and the Apple ecosystem isn’t your jam, this Surface Pro makes a compelling case. Cost and Availability Here’s the catch: it’s not cheap, especially once you add the essential keyboard. The new 12-inch Surface Pro Copilot+ PC starts at $799 (tablet only). If you want that gorgeous Violet (or Ocean), you need to step up to the 512GB storage model, which is $899. Pre-orders are open now, with devices shipping starting May 20. Despite the separate keyboard cost, this sleek, powerful, and finally purple machine hits a sweet spot. Microsoft’s new 12-inch Surface Pro Copilot+ PC in Violet is, without a doubt, my Product of the Week. Microsoft, you finally did it!

In the rapidly expanding world of smart homes, the spotlight has long shone on convenience, connectivity, and compatibility. But as major ecosystems like Amazon Alexa, Google Home, and Apple HomeKit compete for dominance, one initiative — Matter, backed by the Connectivity Standards Alliance (CSA) — is quietly transforming the smart home not only through interoperability but also through an equally vital and underappreciated pillar: security. For Infineon Technologies, Matter’s emphasis on built-in, standardized security isn’t just a feature — it’s central to the company’s long-term strategy. In a recent interview with Infineon’s Steve Hanna, a veteran in IoT security and a leading voice within the Matter initiative, he explained why this often “unsexy” aspect of smart home design may be its most important. Infineon’s Global Strategy and Role in IoT Infineon Technologies is a leading global semiconductor company headquartered in Germany, known for delivering secure, energy-efficient solutions across automotive, industrial, and consumer markets. The company plays a pivotal role in the Internet of Things (IoT) ecosystem by providing embedded security, power management, and connectivity technologies. Infineon’s broad portfolio supports smart home, smart factory, and edge computing applications, enabling reliable and scalable IoT deployments. Its long-term commitment to security and innovation has made it a trusted partner for manufacturers building the next generation of connected devices. Infineon’s overarching corporate strategy is built around two global megatrends: decarbonization and digitalization. The smart home lives squarely at the intersection of these two forces. “Everything is becoming smart,” Hanna said, “not just in the home, but in the workplace, factories, farms — everywhere.” That connectivity can help optimize energy use, reduce waste, and streamline operations. However, it also creates new entry points for cyberattacks, especially as devices become more connected and embedded in daily life. From Infineon’s perspective, it was clear early on that the success of a smart home standard would depend not just on ease of use but on security from day one. So, when the then-Zigbee Alliance began developing a new open standard, Infineon committed to a seat at the table. “When we saw Apple, Amazon, Google, Samsung, and others get involved, we knew this would be big,” Hanna said. “We saw it as the TCP/IP moment for the smart home.” Matter Built Security Into the Standard From Day One Unlike past attempts at unifying the smart home — many of which faltered due to proprietary lock-in or lack of coordination — Matter was built from the ground up on four core principles, one of which is security. That foundational design choice makes Matter fundamentally different. Hanna explained that Matter’s security standards are mandatory, baked in, and non-optional — a significant departure from previous device ecosystems. “Just like your browser won’t access an unsecured site anymore, Matter devices won’t operate without encryption and authentication,” he said. In practical terms, Matter 1.0 includes 10 baseline security features, including secure onboarding, device attestation, encrypted communication, automatic software updates, and more. These aren’t value-adds or premium options — they’re standard requirements for certification. Image courtesy of Infineon Technologies And crucially, manufacturers can’t disable them. Why Security Matters, Even If Consumers Don’t See It Security may not sell smart speakers, but it protects what matters most: consumer trust. Hanna framed the issue in stark terms: “Any connected device becomes a potential attack surface. Lightbulbs, thermostats, locks — if they’re online, they must be secure.” A common vulnerability? Firmware that doesn’t get updated. “Many consumers don’t update their devices,” Hanna noted, “and many manufacturers avoid automatic updates to reduce support costs.” That creates real risk, as attackers routinely target unpatched devices to build botnets or gain access to home networks. Image courtesy of Infineon Technologies Matter solves this with mandatory, secure, seamless, and standardized over-the-air (OTA) updates. The process is transparent to the end user but critical for long-term protection. Even better, the CSA recently introduced a rapid recertification process so manufacturers can deploy critical patches quickly, bypassing long delays that might otherwise expose homes. Certification, Compliance, Build Consumer Confidence To carry the Matter logo, a device must pass rigorous automated testing. Updates must be re-certified by an authorized testing lab or, for trained manufacturers, through an internal automated process that ensures compliance. This framework gives consumers peace of mind but also relieves manufacturers of the burden of customizing devices for each ecosystem. “In the past, developers had to create four different versions of a device to work with Apple, Google, Amazon, and Samsung,” Hanna said. “Matter consolidates that into one universal standard.” That consolidation doesn’t just simplify development — it also ensures a consistent security posture across all certified devices. Infineon’s Role: Security at the Silicon Level As a hardware company, Infineon brings deep expertise in embedded security to the Matter ecosystem. Its secure microcontrollers and cryptographic modules are explicitly designed to support Matter-compliant features, from onboarding to key storage and device attestation. “We’ve been working on IoT security for over 30 years,” said Hanna. “We’re not just implementing standards — we’re helping to shape them.” Infineon’s secure elements are now used in smart locks, lighting, appliances, and more, helping manufacturers implement Matter with minimal friction. Importantly, Infineon products also support advanced features like device identity attestation, enabling ecosystems to detect and block counterfeit or cloned devices — an increasingly serious threat as smart home adoption grows. AI and Revocation Advance Smart Home Security Hanna emphasized that Matter isn’t static. The standard continuously evolves, with new capabilities rolling out each release — a recent example being device revocation. Suppose a cloned or compromised device is detected in the wild. If a cloned or compromised device is detected in the wild, Matter-compliant ecosystems can now block it — a critical protection against supply chain attacks and spoofed devices. Looking ahead, AI adds both complexity and opportunity. “The bad guys are already using AI to find vulnerabilities,” Hanna said. “So, the good guys need to use AI to detect and respond to them faster.” Additionally, as AI becomes embedded in more devices — think learning thermostats or voice-controlled assistants — ensuring the AI is secure becomes another layer of responsibility. Conclusion: Security Is the Killer Feature You Don’t See Matter’s built-in security may be its most disruptive and lasting contribution in a market dominated by hype over compatibility and convenience. It provides the invisible foundation on which trust, privacy, and longevity depend. For companies like Infineon, that’s not just a compliance issue — it’s a competitive advantage. “We believe Matter will do for the smart home what Bluetooth and Wi-Fi did for wireless,” Hanna said. “But it has to be secure from the beginning — and stay secure. That’s what Infineon is committed to delivering.” So, while consumers may never ask if their smart bulb or doorbell has secure firmware updates or cryptographic identity, they’ll benefit from it whenever they turn on the lights, lock the door, or adjust the thermostat. Behind the scenes, Infineon — and the Matter initiative — will ensure those simple moments stay safe, reliable, and smart.

Waymo Builds Arizona Factory To Grow Robotaxi Fleet By John P. Mello Jr. May 7, 2025 5:00 AM PT The 5th-generation Waymo Driver atop an all-electric Jaguar I-Pace SUV, part of Waymo’s current robotaxi fleet. (Source: Waymo) ADVERTISEMENT Rubrik Foward 2025: The Future of Cyber Resilience is here When an attacker comes for your business, will you be ready? Chart your path to cyber resilience and keep your business running. June 4 | Virtual Event | Register Now Waymo, the autonomous vehicle arm of Google’s parent company, Alphabet, revealed Monday that it plans to expand its manufacturing capacity in Arizona to add 2,000 more robotaxis to its fleet. The company explained in a blog that it’s partnering with Magna International, a Canadian auto parts maker, to build a 239,000-square-foot facility in Mesa to retrofit more than 2,000 Jaguar I-Pace electric SUVs with Waymo’s autonomous driving technology. The vehicles will be used to expand Waymo’s robotaxi services to three new metro areas in 2026: Atlanta, Miami, and Washington, D.C. Currently, Waymo has robotaxi operations in Phoenix, San Francisco, Los Angeles, and Austin, Texas. According to Waymo, the new facility’s flexible design will enable it to integrate the 6th-generation Waymo Driver on new vehicle platforms, beginning with the Zeekr RT model later this year. Waymo’s next-gen rider-first robotaxi platform, developed with Zeekr, will debut the 6th-generation Waymo Driver. (Source: Waymo) Waymo explained that the plant will introduce an automated assembly line and other efficiencies over time because it needs to build multiple platforms simultaneously and at higher volumes. It noted that when the facility is operating at full capacity, it will be capable of building tens of thousands of fully autonomous Waymo vehicles per year. “We have made great strides in steadily expanding Waymo One ride-hailing service to multiple commercial markets, and now is the time to scale our technology further to support existing and future deployments,” Waymo Product Communications Manager Chris Bonelli told TechNewsWorld. Significant Announcement “Waymo’s announcement is a significant milestone for the autonomous driving industry,” said Kathleen Rizk, senior director of user experience benchmarking and technology at J.D. Power, a consumer research, data, and analytics firm, in Troy, Mich. “It demonstrates their unwavering commitment to advancing this technology, even as other companies have withdrawn from the market due to various challenges,” she told TechNewsWorld. Edward Sanchez, a senior analyst in the automotive practice of TechInsights, a global technology intelligence company, added that the move has market benefits for Waymo. “This announcement solidifies Waymo’s current position as the leading autonomous ride-hailing provider in the U.S. and signifies the company’s growth ambitions,” he told TechNewsWorld. The commitment to manufacturing could also have broad market benefits, contended Seth Goldstein, an equity strategist and chair of the Electric Vehicle Committee at Morningstar Research Services in Chicago. “Any announcement by a company to build a dedicated robotaxi facility is positive for the broader industry, as it will help accelerate AV [autonomous vehicle] adoption,” he told TechNewsWorld. Sam Abuelsamid, vice president of market research at Telemetry, a Detroit-based transportation research and advisory company, acknowledged that adding additional capacity to up-fit vehicles with the Waymo Driver is a good sign for Waymo’s progress in deploying their robotaxis, but argued it doesn’t really say anything about the rest of the industry at this stage. “Waymo is at least confident enough in the safety of its system and customer acceptance that it is willing to invest in significantly expanding its fleet to support operations in more cities,” he told TechNewsWorld. “Whether anyone else is able to do this anytime soon remains to be seen, but Zoox is likely to be next with its purpose-built robotaxis.” Zoox, which was acquired by Amazon in 2020, offers robotaxi services in the San Francisco Bay Area, Seattle, and Las Vegas. It uses vehicles built explicitly for autonomous ride-hailing rather than retrofitting existing vehicles for that purpose. Market Leader Abuelsamid noted that in North America, Waymo is the clear leader in developing robust automated driving systems (ADS) and the back-end services required to run an operation. They’ve also developed fairly healthy relationships with local authorities to integrate the systems with first responders, get curb access, and overcome other challenges. “Zoox is likely to be next in line to deploy commercially,” he said. “Tesla aims to launch a service in Austin next month, but I don’t see evidence that it’s close to being able to safely operate without supervision yet. My guess is they will be operating with a remote driver supervisor on each car to be ready to disengage at any moment, so it would technically still be a Level 2 system, like what they have today.” A Level 2 autonomous driving vehicle is equipped with advanced driver assistance systems (ADAS) that simultaneously control both steering and acceleration/deceleration. However, the human driver must remain engaged and ready to take over at any time. Morningstar technology equity analyst Malik Ahmed Khan noted that Waymo is quite well-positioned in the competitive autonomous vehicle market. “It is the only paid AV taxi service right now doing more than 250,000 rides a week,” he told TechNewsWorld. Khan explained that Waymo’s business model is quite scalable, especially as the firm’s Hyundai/Zeekr vehicles help drive down costs. Interior view of Waymo’s rider-first vehicle platform, designed for autonomous ride-hailing in partnership with Zeekr. (Source: Waymo) Safety Still a Concern for AV Adoption It also differentiates itself from competitors in a number of ways, he added. “Certainly, there is a first-mover/brand awareness differentiation,” Khan observed. Waymo also has deep pockets. “Having a well-capitalized backer such as Alphabet helps, especially when it comes to differentiation against smaller companies with limited access to capital,” Khan said. “Word of mouth and perceptions of safety matter, and Waymo’s strong penetration into its markets shows that an increasing number of riders are on board Waymo from a safety perspective,” he added. Rizk noted that according to J.D. Power’s 2024 U.S. Robotaxi Experience Study, safety concerns are a significant obstacle to consumer acceptance of autonomous driving. “Consumers want to see federal and state regulations to support this industry,” she said. “Furthermore,” she continued, “safety concerns impact consumer comfort with fully automated self-driving vehicles, and our study findings indicate that 83% of consumers desire safety statistics about the technology before using autonomous vehicles. These safety statistics should not only include data but also demonstrate a proven track record over time, validating consistent performance.” She added that consumers also have concerns that automated vehicle technology may negatively affect data privacy and be susceptible to hacking, as over 80% want to know what measures are being taken to prevent the hacking of AVs. Mainstream AV Use Still a Decade Away With its investment in new manufacturing capacity, Waymo clearly has its eyes focused on a mainstream future. However, that future may be slow to develop. Vehicle deployment, for instance, will be spotty. “The pace of deployment for autonomous vehicles varies widely by region,” TechInsights’ Sanchez said. “In the U.S., it is largely a state-by-state proposition in terms of approval, whereas in markets such as China with more centralized control, approval can be granted on a country-wide basis.” “Even in the EU,” he added, “there are still regional restrictions, although the European Commission is looking to streamline the approval process to allow for multi-country operation of autonomous vehicles.” Morningstar forecasts autonomous vehicles will see rapid adoption growth over the next decade and make up 50% of all ride-hailing rides by 2035 in the U.S. and Canada. “We expect AVs will roll out in a city-by-city approach, the same way Uber and Lyft did last decade,” Goldstein said. “Our bullish outlook is driven by our view that AVs will be cheaper than traditional ride-hailing rides, leading to higher adoption,” he continued. “We also see barriers to adoption falling in the coming years to drive higher growth.” Telemetry’s Abuelsamid also doesn’t expect autonomous vehicles to become common before the end of the decade. “While strides have been made on the hardware cost, the operational costs of a robotaxi system are still very high, and not even Waymo is close to profitability,” he explained. “There is also the problem of utilization because robotaxi demand is uneven during the day, so finding a way to use the vehicles for tasks like deliveries during low-demand periods will be essential,” he said. “I think applications like robo-shuttles to support and expand transit services, long-haul trucking, and middle-mile deliveries make more sense from a business perspective in the near term.” John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. Related Stories More by John P. Mello Jr. view all More in Transportation

Chatbots Having Minimal Impact on Search Engine Traffic: Study By John P. Mello Jr. May 6, 2025 5:00 AM PT ADVERTISEMENT Secure Mobile Payments: Protect Wallets and SoftPOS From Growing Cyber Threats Learn how attackers exploit digital wallets and SoftPOS -- and how you can defend against them. Join our technical webinar on May 13. Reserve your spot today! AI chatbots have barely made a dent in traffic to popular search engine sites over the past two years, according to a study by SEO and backlink services firm OneLittleWeb. The study analyzed global web traffic from April 2023 to March 2025. In the most recent year, chatbot sites accounted for just 2.96% of the visits received by search engines. Between April 2024 and March 2025, search engine traffic declined only slightly — down 0.51% to 1.86 trillion visits — while chatbots saw an 80.92% year-over-year spike in traffic. The modest drop in search traffic suggests that, despite explosive growth, AI chatbots are not yet displacing traditional search behavior in any meaningful way. “Even with ChatGPT’s massive growth, it still sees approximately 26 times fewer daily visits than Google,” wrote the author of the study, Sujan Sarkar, founder of OneLittleWeb. “While AI chatbots are growing fast, search engines continue to hold a dominant position in daily user engagement,” he added. Sarkar also noted that search engines like Google and Microsoft Bing are leveraging AI features like AI Overviews and Search Generative Experience (SGE) to increase traffic in early 2025. “Despite a temporary dip, these integrations have helped revive interest and usage,” he wrote. Why Search Still Outpaces AI Chatbots The length of the study period could be distorting the current chatbot versus search engine picture, contended Rob Enderle, president and principal analyst of the Enderle Group, an advisory services firm in Bend, Ore. “Using AI for search is relatively new, and people haven’t learned how to properly prompt yet, so a survey looking back over two years of data would reflect a world mostly pre-AI search and one where the tools were just starting to be used at the end of the survey period, biasing the results toward poor AI use,” he told TechNewsWorld. “With emerging technologies, surveys need to be more real-time, as the old data won’t reflect current reality,” he said. Search engines also have advantages for drumming up traffic. “Search engines benefit from decades of user trust and deeply ingrained habits,” explained Mark N. Vena, president and principal analyst at SmartTech Research in Las Vegas. “They offer a fast, comprehensive way to find information, products, news, and more — across virtually every category,” he told TechNewsWorld. “Their massive web indexing and ad-supported business models fund constant innovation.” He added, “Integration with browsers and mobile devices as the default search method reinforces daily usage.” Evolution, Not Dissolution The study also maintained that search engines are evolving rather than fading, integrating AI tools to offer a richer, more personalized user experience. At the same time, chatbots are carving out their niche in tasks requiring direct, customized responses. “Search engines are not disappearing; they’re adapting by embedding AI to improve response quality and personalization,” Vena said. “Meanwhile, chatbots thrive in scenarios requiring synthesis, dialogue, or creative output.” “The tools serve different intents,” he continued. “Search is broad and navigational. Chatbots are task-focused and conversational. Their coexistence reflects a diversification of how users seek and process information.” Enderle, though, argued that AI providers are struggling to carve out their niche. “This will change over time, but I’d have thought it would happen sooner than it is happening. I blame the lack of promotion as the cause for the slow pickup,” he said. “There’s some convergence of AI and search happening, as Google integrates more AI and ChatGPT emphasizes and improves search,” added Greg Sterling, co-founder of Near Media, a market research firm in San Francisco. “Search engines aren’t dying, but the old user experience is, with implications for publishers and advertisers,” he told TechNewsWorld. Chatbots Expand Into Companionship, Therapy Alex Ambrose, a policy analyst with the Information Technology and Innovation Foundation, a research and public policy organization in Washington, D.C., agreed that chatbots are attracting users with personalized experiences. “Users are increasingly turning to chatbots not just for tasks like rebooking flights or automating workflows but for companionship and therapeutic applications,” she told TechNewsWorld. “These chatbots — or AI companions — are taking the technology a step further by providing users with extremely personal experiences in the shape of providing emotional support and digging into deeper personal questions,” she said. The study also ranked chatbots by visits. ChatGPT was at the top of the list, followed by DeepSeek, Gemini, Perplexity, Claude, Microsoft Copilot, Blackbox AI, Grok, Monica, and Meta AI. It noted the fastest-growing chatbots were DeepSeek and Grok. DeepSeek experienced a staggering surge in traffic, with total visits jumping from 1.5 million to 1.7 billion during the two-year study period, an increase of 113,007%. Grok’s growth was 353,787%, increasing from 61,200 visits to 216.5 million. “Grok and DeepSeek are growing fast due to strong backing and strategic positioning,” Vena said. “Grok leverages its integration into X — formerly Twitter — giving it immediate reach and visibility,” he explained. “DeepSeek appeals to users looking for more technical, research-oriented AI capabilities. Both benefit from user curiosity about alternatives to ChatGPT and differentiated conversational experiences.” Chatbots vs. Search: The Battle for User Intent Vena contended that the real contest isn’t just about traffic. “It’s about controlling the user’s starting point when they have a question or goal,” he said. “Chatbots may win in productivity or assistance, while search engines still dominate for broad exploration and commerce. Integration and default positioning will shape the future more than features alone. The next wave may involve blended experiences that merge the strengths of both.” Sterling agreed that the simple traffic analysis approach doesn’t tell the whole story about how usage is changing. “As people become more sophisticated about AI, they’re being more discriminating about how to use it versus search,” he noted. “The idea that people either use AI or search is false. Both are being used, but the ways that AI and search are used are evolving.” “More and more research will migrate to AI unless Google co-opts that,” he continued. “Google will probably hold on to most navigational, shopping, and commercial queries.” “Having said that,” Sterling added, “the market is now moving very fast and evolving quickly. Any prediction made today may turn out to be wrong tomorrow.” Enderle pointed out that the market is at the very beginning of this trend. “I expect by 2030 kids will look back at non-AI search engines like they now look back at dial phones, asking, how anyone lived in these dark times,” he predicted. John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. Related Stories More by John P. Mello Jr. view all More in Search Tech

A wave of disruption is sweeping through AI. Meta’s recent unveiling at LlamaCon 2025 of the roadmap for its Llama family of large language models (LLMs) paints a compelling picture, one where open source isn’t just a preference, but the very engine driving AI’s future. If Meta’s vision comes to fruition, we’re not just looking at incremental improvements; we’re facing an AI tsunami powered by collaboration and accessibility, threatening to wash away the walled gardens of proprietary models. Llama 4: Faster, Multilingual, Vast Context The headline act, Llama 4, promises a quantum leap in capabilities. Speed is paramount, and Meta claims significant acceleration, making interactions feel more fluid and less like waiting for a digital oracle to deliver its pronouncements. But the true game-changer appears to be its multilingual prowess, boasting fluency in a staggering 200 languages. Imagine a world where language barriers in AI interactions become a quaint historical footnote. This level of inclusivity has the potential to democratize access to AI on a truly global scale, connecting individuals regardless of their native tongue. Furthermore, Llama 4 is set to tackle one of the persistent challenges of LLMs: context window limitations. The ability to feed vast amounts of information into the model is crucial for complex tasks, and Meta’s claim of a context window potentially as large as the entire U.S. tax code is mind-boggling. Think of the possibilities for nuanced understanding and comprehensive analysis. The dreaded “needle in a haystack” problem — retrieving specific information from a large document — is also reportedly seeing significant performance improvements, with Meta actively focused on making it even more efficient. This enhanced ability to process and recall information accurately will be critical for real-world applications. Scalability Across Hardware Meta’s strategy isn’t just about building behemoth models; it’s also about making AI accessible across a range of hardware. The Llama 4 family is designed with scalability in mind. “Scout,” the smallest variant, is reportedly capable of running on a single Nvidia H100 GPU, making powerful AI more attainable for individual researchers and smaller organizations. “Maverick,” the mid-sized model, will also operate on a single GPU host, striking a balance between power and accessibility. While the aptly named “Behemoth” will undoubtedly be a massive undertaking, emphasizing smaller yet highly capable models signals a pragmatic approach to widespread adoption. Crucially, Meta touts a very low cost-per-token and performance that often exceeds other leading models, directly addressing the economic barriers to AI adoption. Llama in Real Life: Diverse Applications Llama’s reach extends beyond earthly confines. Its deployment on the International Space Station, providing critical answers without a live connection to Earth, highlights the model’s robustness and reliability in extreme conditions. Back on our planet, real-world applications are already transformative. Sofya, a medical application leveraging Llama, is substantially reducing doctor time and effort, promising to alleviate burdens on healthcare professionals. Kavak, a used car marketplace, is using Llama to provide more informed guidance to buyers, enhancing the consumer experience. Even AT&T is utilizing Llama to prioritize tasks for its internal developers, boosting efficiency within a major corporation. A partnership between Box and IBM, built on Llama, further assures both performance and the crucial element of security for enterprise users. Open, Low-Cost, User-Centric AI Meta aims to make Llama fast, affordable, and open — giving users control over their data and AI future. The release of an API to improve usability is a significant step towards this goal, lowering the barrier to entry for developers. The Llama 4 API promises an incredibly user-friendly experience, allowing users to upload their training data, receive status updates, and generate custom fine-tuned models that can then be run on their preferred AI platform. This level of flexibility and control is a direct challenge to the closed-off nature of some proprietary AI offerings. Tech Upgrades and Community Enhancements Technological advancements are furthering Llama’s capabilities. Implementing speculative decoding reportedly improves token generation speed by around 1.5x, making the models even more efficient. Because Llama is open, the broader AI community is actively contributing to its optimization, with companies like Cerebras and Groq developing their own hardware-specific enhancements. Llama Adds Powerful Visual AI Tools The future of AI, according to Meta, is increasingly visual. The announcement of Locate 3D — a tool that identifies objects from text queries — and continued development of the Segment Anything Model (SAM) — a one-click tool for object segmentation, identification, and tracking — signal a shift toward AI that can truly “see” and understand the world around it. SAM 3, launching this summer with AWS as the initial host, promises even more advanced visual understanding. One highlighted application is the ability to automatically identify all the potholes in a city, showcasing the potential for AI to address real-world urban challenges. Conversational AI in Action Llama’s user-friendly design is already translating into meaningful real-world applications. Comments from Mark Zuckerberg and Ali Ghodsi of Databricks reinforced the shift toward smaller yet more powerful models, accelerated by rapid innovation. Even traditionally complex tools like Bloomberg terminals now respond to natural language queries, eliminating the need for specialized coding. The real-world impact is already evident: the Crisis Text Line uses Llama to assess risk levels in incoming messages — potentially saving lives. Open Source Advantages and Future Challenges Ali Ghodsi emphasized Databricks’ belief in open source, citing its ability to foster innovation, reduce costs, and drive adoption. He also highlighted the growing success of smaller, distilled models that increasingly rival their larger counterparts in performance. The anticipated release of “Little Llama” — an even more compact version than Scout — further underscores the momentum behind this trend. Looking ahead, the focus shifts to safe and secure model distillation — ensuring smaller models don’t inherit vulnerabilities from their larger predecessors. Tools like Llama Guard are early steps in addressing these risks, but more work is needed to maintain quality and security across a growing range of models. One emerging concern is objectivity: open models may recommend a competitor’s product if it’s genuinely the best fit, potentially leading to more honest and user-centric AI. Ultimately, while AI capabilities are advancing rapidly, the real competitive edge lies in data. Encouragingly, as models become more capable, the skills needed to work with them are becoming more accessible. Wrapping Up: Open Source AI’s Rising Power Meta’s Llama 2025 roadmap signals a decisive shift towards open source as the dominant paradigm in AI development. With faster, more multilingual models, a focus on accessibility across various hardware, and a commitment to user control, Meta is unleashing an AI tsunami that promises to democratize the technology and drive unprecedented innovation across industries. The emphasis on real-world applications, from healthcare to education to everyday interactions, underscores the transformative potential of this open and collaborative future of artificial intelligence.

Over several decades as an industry analyst, I have followed the evolution of telecom, wireless, broadband, pay TV, and more. Today, a new transformation is taking shape. Artificial intelligence and 5G wireless — both part of the broader telecom ecosystem — are now converging to create intelligent, self-optimizing networks that are reshaping industries from telecommunications to health care. Innovation has always driven change across every industry, but the pace is accelerating with the rise of technologies like artificial intelligence. Different forms of AI are proliferating, just as 5G and wireless networks have already transformed how we live and work. Now, these forces are beginning to converge — and the impact will be profound. Understanding how AI and 5G are coming together and what this means for customers, workers, investors, and executives is critical as we move into a new era of accelerated change. Let’s pull the camera back for a longer-term, historical perspective on the forces rewriting everything around us — and why companies must move quickly to keep up. AI, 5G, Wireless, and Telecom Are Merging The blending of AI, 5G, wireless, and telecom is already reshaping industries — but what exactly does it mean, and what changes are occurring today and coming next? While different forms of artificial intelligence have existed for decades, the sector has advanced rapidly over the past two years, accelerated by the launch of OpenAI’s ChatGPT and other generative AI technologies. It’s important to understand that AI is not just one type of technology. It is an umbrella term covering different categories and applications. This is where much of the confusion begins. Think of AI as a top-level catchphrase. Underneath it are different types, such as Narrow AI (task-specific intelligence), the still-theoretical General AI (broad human-like intelligence), and the future concept of Super AI (intelligence beyond human capability). Within Narrow AI, we find applications like ChatGPT and other forms of Generative AI, which are rapidly transforming business and consumer experiences. AI’s Growing Impact Across Connectivity Sectors Artificial intelligence is not a single technology but a broad term covering many types, each affecting different industries and companies in distinct ways. Understanding these differences is critical because AI is evolving rapidly and rewriting the rules across every sector it touches. Some companies are acting quickly to integrate AI. These early movers gain a first-to-market advantage. Others are taking a wait-and-see approach, positioning themselves as fast followers — ready to act once the industry’s direction becomes clearer. Both strategies — leading and fast following — can succeed, but they represent very different paths to growth. We’ve seen a similar dynamic before in the wireless industry. AT&T, for example, is often first to market with new technologies — sometimes right, sometimes wrong. Verizon tends to move in once the path is better defined, succeeding as a fast follower. AI Is Driving New Growth Opportunities Artificial intelligence gives companies a new opportunity to reverse a decline and drive growth. AI is already transforming network operations in wireless, telecom, and broadband. Traditionally, service disruptions triggered a scramble to locate, isolate, and fix the problem before customers noticed. Today, AI can detect potential issues early, reroute traffic automatically, and keep networks running smoothly, often without noticeable disruption. This shift is more than a technical upgrade; it’s a competitive advantage. Companies integrating AI into their operations can deliver better service, improve customer satisfaction, and position themselves for renewed growth. While we are still in the early stages of the AI revolution, the companies that move first will be better positioned to shape the next phase of industry transformation. Leaders Emerging in the AI and 5G Race The race to lead the next wave of AI and 5G innovation is already underway — and it’s crowded. Major telecom carriers like AT&T, T-Mobile, Verizon, and Comcast are racing to integrate AI and 5G into their networks. Infrastructure providers like Cisco, Nokia, Ericsson, and Qualcomm are building the technology backbone to support these shifts. Meanwhile, device makers such as Apple, Google, Samsung, and Netgear are embedding AI into products that will drive consumer adoption. As the landscape evolves, we can expect a mix of strategies. Some companies will remain independent, while others will form partnerships or pursue mergers and acquisitions to strengthen their positions. However, leadership in the AI era won’t be limited to the largest companies. Even smaller players are finding ways to lead by moving quickly with AI-driven innovation. For instance, RedChip recently introduced RedChat, an AI-based investment service designed to help investors evaluate and select small-cap stocks. This offers a fresh example of how AI is opening new growth paths across industries, not just in telecom and wireless. In this rapidly shifting environment, size alone won’t determine success. The next generation of winners will be defined by agility, innovation, and strategic execution. The Urgent Race to Master AI The AI landscape is changing rapidly — and no one fully understands the entire picture. Most companies specialize in one aspect of AI, but few grasp the full scope of what’s unfolding. It’s a confusing but opportunity-filled moment for those who choose the right path forward. Just two years ago, even senior executives at leading companies underestimated how quickly generative AI tools like ChatGPT would reshape the technology landscape — and some even blocked employees from using them. We know far more today than we did two years ago — and far less than we will tomorrow. That’s the excitement and the risk AI brings: new opportunities, new dangers, and a dramatically faster pace of change. Executives who move quickly will define the future — those who hesitate risk being left behind.

The rapid proliferation of artificial intelligence is both exhilarating and deeply concerning. The sheer power unleashed by these algorithms, largely concentrated within the coffers and control of a handful of tech behemoths — you know, the usual suspects, the ones who probably know what you had for breakfast — has ignited a global debate about the future of innovation, fairness, and even societal well-being. The ongoing scrutiny and the looming specter of regulatory intervention are not merely bureaucratic hurdles; they are a necessary reckoning with the profound risks inherent in unchecked AI dominance. It’s like we’ve given a few toddlers the keys to a nuclear-powered Lego set, and now we’re all nervously watching to see what they build (or break). Let’s talk about how AI algorithms are reshaping society, who controls them, and why the stakes are far higher than most people realize. Then, we’ll close with my Product of the Week: a new Wacom tablet I use to put my real signature on digital documents. Bias Risks in AI: Intentional and Unintentional The concentration of AI development and deployment within a few powerful tech companies creates a fertile ground for the insidious growth of both intentional and unintentional bias. Intentional bias, though perhaps less overt (think of it as a subtle nudge in the algorithm’s elbow), can creep into the design and training of AI models when the creators’ perspectives or agendas, whether conscious or subconscious, shape the data and algorithms. This can manifest in subtle ways, prioritizing certain demographics or viewpoints while marginalizing others. For instance, if the teams building these models lack diversity, their lived experiences and perspectives might inadvertently lead to skewed outcomes. It’s like asking a room full of cats to design the perfect dog toy. However, the more pervasive and perhaps more dangerous threat lies in unintentional bias. AI models learn from the data they are fed. If that data reflects existing societal inequalities (because humanity has a history of not being entirely fair), AI will inevitably perpetuate and even amplify those biases. Facial recognition software, notoriously less accurate for individuals with darker skin tones, is a stark example of how historical and societal biases embedded in training data can lead to discriminatory outcomes in real-world applications, from law enforcement to everyday convenience. The sheer scale at which these dominant tech companies deploy their AI systems means these biases can have far-reaching and detrimental consequences, impacting access to opportunities, fair treatment, and even fundamental rights. It’s like teaching a parrot to repeat all the worst things you’ve ever heard. Haste Makes Waste, Especially When Algorithms Are Involved Adding to these concerns is the relentless pressure within these tech giants to prioritize productivity and rapid deployment over the crucial considerations of quality and accuracy. In the competitive race to be the first to market with the latest AI-powered feature or service (because who wants to be the Blockbuster of the AI era?), the rigorous testing, validation, and refinement processes essential to ensuring reliable and trustworthy AI are often sidelined. The “move fast and break things” ethos, while perhaps acceptable in earlier stages of software development, carries significantly higher stakes when applied to AI systems that increasingly influence critical aspects of our lives. It’s like releasing a self-driving car that’s only been tested in a parking lot. The consequences of prioritizing speed over accuracy can be severe. Imagine an AI-powered medical diagnosis tool that misdiagnoses patients due to insufficient training on diverse datasets or inadequate validation, leading to delayed or incorrect treatment. Or consider an AI-powered hiring algorithm that, optimized for speed and volume, systematically filters out qualified candidates from underrepresented groups based on biased training data. The drive for increased productivity, fueled by the immense resources and market pressure these dominant tech companies face, risks creating an ecosystem of AI that is efficient but fundamentally flawed and potentially harmful. It’s like trying to win a race with a car that has square wheels. Ethical Oversight Lags in AI Governance Perhaps the most alarming aspect of the current AI landscape is the relative lack of robust ethical oversight within these powerful tech organizations. While many companies espouse ethical AI principles (usually found somewhere on page 78 of their terms of service), implementing and enforcing these principles often lag far behind the rapid advancements in the technology itself. The decision-making processes within these companies regarding the development, deployment, and governance of AI systems are often opaque, lacking independent scrutiny or clear mechanisms for accountability. The absence of strong ethical frameworks and independent oversight creates a vacuum where potentially harmful AI applications can be developed and deployed without adequately considering their societal impact. The pressure to innovate and monetize AI can easily overshadow ethical considerations, allowing harmful outcomes — such as bias, privacy violations, or erosion of human autonomy — to go unaddressed until after damage is already done. The sheer scale and influence of these dominant tech companies necessitate a far more rigorous and transparent approach to ethical AI governance. It’s like letting a toddler paint the Mona Lisa. The results are likely to be abstract and possibly involve glitter. Building a Responsible AI Future The risks inherent in the unchecked dominance of AI by a few large tech companies are too significant to ignore. A multi-pronged approach is needed to foster a more responsible and equitable AI ecosystem. Stronger regulation is a critical starting point. Governments must move beyond aspirational guidelines and establish clear, enforceable rules that directly address the risks posed by AI — bias, opacity, and harm among them. High-stakes systems should face rigorous validation, and companies must be held accountable for the consequences of flawed or discriminatory algorithms. Much like the GDPR shaped data privacy norms, new legislation — call it AI-PRL, for AI Principles and Rights Legislation — should enshrine basic protections in algorithmic decision-making. Open-source AI development is another key pillar. Encouraging community-driven innovation through platforms like AMD’s ROCm helps break the grip of closed ecosystems. With the proper support, open AI projects can democratize development, enhance transparency, and broaden who gets a say in AI’s direction — like opening the recipe book to every cook in the kitchen. Fostering independent ethical oversight is paramount. Creating ethics boards with the authority to audit and advise on AI deployment — particularly at dominant firms — can introduce meaningful checks. Drawing from diverse disciplines, these bodies would help companies uphold ethical standards rather than self-regulate in the shadows. Think of them as the conscience of the industry. Mandating transparency and explainability in AI algorithms is essential for building trust and enabling accountability. Users and regulators alike need to understand how AI systems arrive at their decisions, particularly in high-stakes contexts. Requiring companies to provide clear and accessible explanations of their algorithms while protecting legitimate trade secrets can help identify and address potential biases and errors. It’s like asking the Magic 8 Ball to show its workings. Finally, investing in AI literacy and public education is crucial for empowering individuals to understand AI’s capabilities and limitations, as well as its potential risks and benefits. A more informed public will be better equipped to engage in the societal debates surrounding AI and demand greater accountability from the companies that develop and deploy these powerful technologies. Wrapping Up: Charting a Course for Responsible AI The algorithmic tightrope we are currently walking demands careful and deliberate steps. The immense potential of AI must be harnessed responsibly, with a keen awareness of the risks inherent in unchecked power. By implementing robust regulations, fostering open-source alternatives, mandating ethical oversight and transparency, and investing in public education, we can strive towards an AI ecosystem that benefits all of society, rather than exacerbating existing inequalities and concentrating power in the hands of a few. The future of AI, and indeed a significant part of our own future, depends on our collective willingness to navigate this algorithmic tightrope with wisdom, foresight, and a commitment to ethical innovation. One by Wacom Image Credit: Wacom In a world increasingly dominated by digital documents, the simple act of a handwritten signature can feel like a quaint relic. Enter the One by Wacom small graphics tablet, a surprisingly affordable portal to bridge the analog and digital, especially for those of us (read: me) tired of fumbling with a mouse to create a digital scrawl that vaguely resembles our John Hancock. Priced at a tempting $39.94 on Amazon (for the wired version), this little slate offers a far more natural way to “sign here” without resorting to pre-saved images that lack that personal touch. For my primary use case — imprinting my actual, legible (virtually none of the time) signature onto digital contracts and forms — One by Wacom is a genuine game-changer. Gone are the jagged lines and shaky approximations of my name. Instead, the pressure-sensitive pen glides smoothly across the tablet’s surface, translating my familiar loops and flourishes onto the screen with surprising accuracy. Interestingly, Wacom offers this petite powerhouse in wired and Bluetooth wireless flavors. The wireless version, while liberating from the tyranny of cables, will set you back a heftier $79.94. While the freedom of a wireless setup is alluring, especially for cluttered desks (guilty!), the wired version arguably presents a better value proposition. No charging anxieties, no pairing woes — just plug it in and get signing. For a tool primarily used for quick tasks like signatures, the tethered existence feels like a small price to pay for perpetual power. But the One by Wacom is more than a fancy digital autograph machine. This versatile gadget opens up a surprising array of creative possibilities. Aspiring digital artists can use it for basic sketching and drawing, enjoying a more intuitive experience than a mouse allows. Photo editors can leverage the pen’s pressure sensitivity for more precise retouching and masking. Even navigating your computer can become a slightly more artistic endeavor, though perhaps less efficient than a traditional mouse for everyday tasks. Think of it as adding a touch of flair to the mundane. While the “small” in its product description is accurate, the active drawing area is perfectly adequate for signatures and basic creative work. It’s portable enough to tuck into a laptop bag, making it a handy tool for on-the-go professionals who need to sign documents remotely or sketch ideas wherever inspiration strikes. The One by Wacom small tablet is a surprisingly capable and affordable tool. For anyone seeking a more natural way to sign digital documents, the wired version at under $40 is a no-brainer. While it might not replace a dedicated graphics tablet for serious artists, its versatility extends to basic drawing and photo editing, offering a fun and intuitive alternative to the humble mouse. It’s a small investment that can make a big difference in your digital workflow — finally allowing your signature to have the personality it deserves, even in the cold, hard world of cyberspace — making the One by Wacom tablet my Product of the Week.

Cybersecurity threats are escalating, yet many small and medium-sized businesses (SMBs) remain dangerously unprotected. Experts point to a combination of high costs, technical complexity, and a shortage of skilled cybersecurity professionals as key reasons why SMBs often delay or avoid adopting critical protections. Even businesses willing to invest face challenges navigating a crowded and confusing marketplace of tools and services. This lack of preparedness has made SMBs increasingly attractive targets for cybercriminals seeking easy access to valuable data and networks. In September, the FBI, with intelligence support from cybersecurity researchers, disrupted a major Chinese botnet that infected more than 200,000 consumer devices worldwide — a reminder of how rapidly cyberthreats are evolving. Attacks against businesses of all sizes, including SMBs, have surged in recent years, increasing the stakes for those without strong defenses. Several factors have contributed to the growing cybersecurity risks facing SMBs. The swift transition to remote work expanded the number of attack surfaces. Digital transformation efforts and large-scale cloud service adoption by smaller firms have also created larger security gaps. Increased reliance on third-party vendors and supply chains has further amplified these vulnerabilities, observed Jerry Chen, former Cisco engineer and co-founder of cybersecurity firm Firewalla. “The uptick in attacks against SMBs is part of a broader trend where cybercriminals focus on relatively unprotected devices to build large-scale botnets capable of launching distributed denial-of-service [DDoS] attacks,” he told TechNewsWorld. Cyberattacks Against SMBs Surge Chen noted that small business networks are desirable targets for attackers because they are especially vulnerable. They often lack the budget to protect themselves with effective security tools and the resources to hire appropriate cybersecurity talent. Protection tradeoffs are a necessary reality for SMBs, he noted. Smaller businesses can’t have the same cybersecurity measures as enterprises. Those systems are costly and cannot be operated without an IT team to manage them. SMBs need cost-effective solutions that bring visibility to their networks. That means a solution that can always alert them to how many devices are on their network, what they are doing, and whether they are transferring any sensitive data. “Maintaining this kind of vigilance is the only way to know if there is any risk,” Chen asserted. Steve Garrison, SVP of marketing at Stellar Cyber, agrees that hackers realize there are millions of companies in the SMB and mid-market small and medium-sized enterprise (SME) space. “There is less sophistication in terms of cyber awareness in the SMB space. So bad people are just starting to cash in on it,” he told TechNewsWorld. It’s a numbers game, and it’s easier than ever for hackers to use your devices as a door to get your or the company’s personal information. Nobody provides a desktop phone anymore, Garrison noted. SMBs Struggle With Outdated Cyber Tools When businesses struggle with cyber resources, they often adopt outdated technology or basic security measures that leave them vulnerable. Most small business operators do not know how to implement best practices. “This can lead to misconfigurations, weak passwords, unpatched software, or poor incident response planning, all of which increase the risk of a breach,” said Chen. This situation plays directly into hackers’ hands. Cybercriminals are conscious of all these factors and target small businesses precisely because they are seen as easy targets. “Small businesses may not believe they are at risk. The ‘it won’t happen to me’ mindset makes them less likely to adopt strong defenses or even implement basic security measures,” Chen offered. “For them, the best solution is making use of tools that are within their budget, effective, and also very simple to work with.” For example, SMBs can use a firewall device that provides comprehensive visibility of the devices operating within their network, their continuously updated activity, and gives them the ability to manage and group them. Not an outdated cyber tool, firewalls with those features can be a good stopgap measure. Basic Cybersecurity Practices for SMBs Chen added that budgetary deficiencies and absent IT workers do not mean companies are defenseless. A system scan tool can investigate the security of an SMB’s network for commonly used ports and vulnerabilities. “An effective scan tool will detect issues such as services that do not have password protection or services that may have a default password or a common password that would be simple for a hacker to guess,” he said. Many SMBs fail to take other basic security steps. On the network visibility side, they should set up controls or rules within their network, such as segmenting essential devices, like work laptops and security cameras, from less critical ones, like personal phones and guest devices. They should also consider isolating IoT devices, which are common points of entry for hackers, to reduce the risk of security breaches or applying an extra layer of protection to guest devices. Other steps include blocking traffic from certain countries, geographic regions, and dangerous websites from accessing the network. Two basic systems help with these tasks: Intrusion Detection Systems (IDS) monitor network traffic and alert administrators to suspicious activity, vulnerabilities, or policy violations, but they do not block threats directly. Intrusion Prevention Systems (IPS) detect and automatically block malicious traffic in real time, preventing harmful connections from reaching the network or causing damage. Combining IDS and IPS strengthens network security by pairing real-time alerts with automatic threat blocking, minimizing risks without heavy hands-on management. “On the device side, SMBs should always make sure that they have the latest firmware updates and fixes ready to implement and use devices that are actively supported by their manufacturers,” Chen urged. Hidden Cybersecurity Features SMBs Can Leverage Some cybersecurity features are built into platforms but may not be evident to buyers. Communications and networking platforms may already have built-in cyber protections, eliminating the need to add third-party software, noted Stellar Cyber’s Garrison. Its Open XDR platform is a good example. The company does not market to individual SMBs. It delivers cyber protection through platform providers. XDR (Extended Detection and Response) systems are evolving technologies that unify threat prevention, detection, and response capabilities. According to Garrison, Stellar Cyber’s Open XDR emphasizes open functionality rather than relying solely on open-source components. It uses a combination of open hooks, webhooks, APIs, connectors, and parsers to link tools the customer has already purchased to ingest companies’ data and fill in the rest of the attack surface for what they do not have. “It’s a one-size-fits-all platform,” he added. “It is a different strategy than your larger [cyber] companies would offer. So it’s another way we go to market to make it more amenable to adopt the technology.” Affordable Next-Gen Firewalls for SMBs Firewalla offers a series of enhanced firewall and router devices designed to protect networks and devices at home and at work. One of the newer models, the Firewalla Gold Pro, can route and inspect network traffic at 10-gigabit speeds and supports Wi-Fi 7, making it well-suited for the fastest networks in small businesses and homes. The Firewalla Gold Pro is a multi-gigabit firewall that is easy to install, simple to use, and requires no monthly fees. It is operated through a smartphone app with various features, such as a vulnerability scanner, an IDS/IPS system to detect and block unauthorized access attempts, and tools for managing and grouping connected devices across the network. Chen said the Firewalla Gold Pro provides the same high-quality performance as enterprise-level security solutions but offers more value due to its cost-effectiveness and technical accessibility. “It is ideal for not only SMBs looking to gain basic visibility, control, and protection of their networks but also for SMBs upgrading their infrastructures and adopting next-generation networks like Wi-Fi 7 for faster speeds,” he added. The Firewalla Gold Pro device is powered by a quad-core Intel processor and 8 gigabytes of RAM, allowing it to scale with growing network demands. Its 10-gigabit ports can be configured for wide-area networks (WANs) or local area networks (LANs). Users can segment traffic using virtual local area networks (VLANs), separating devices into groups for better management and security, all operating at full 10-gigabit speeds. One port can connect to a 10-gigabit Wi-Fi 7 access point and another to a high-speed switch.

Google AI Overviews Hurting Click-Through Rates: Studies By John P. Mello Jr. April 23, 2025 5:00 AM PT Studies show Google's AI Overviews reduce clicks to websites by answering queries directly, raising concerns over traffic and visibility. ADVERTISEMENT Proven Tactics to Scale SMB Software Companies in Competitive Markets Gain market share, boost customer acquisition, and improve operational strength. Get the SMB Software Playbook for Expansion & Growth now -- essential reading for growing tech firms. Free Download. Two studies released last week indicate that Google’s AI Overviews are having a negative impact on click-through rates from online searches, which could ultimately reduce traffic to original sources and affect the quality of content on the internet. An analysis of 300,000 keywords by Xibeijia Guan, a data scientist at SEO tool provider Ahrefs, found that the presence of an AI Overview in search results correlated to a 34.5% lower average click-through rate (CTR) for the top-ranking page compared to keywords without an Overview. “This isn’t surprising. I have seen anecdata suggesting that some websites have seen clicks reduce by 20% to 40% since the rollout of AI overviews,” Ryan Law wrote in an Ahrefs blog on April 17. He explained that AI Overviews function in a similar way to Featured Snippets. They try to resolve the searcher’s query directly, which likely contributes to more zero-click searches. “And although AI Overviews often contain citation links, there can be many of these links cited, making it less likely for any single link to earn the lion’s share of clicks,” he wrote. “Assuming AI Overviews stay in this current form, this is also likely the highest the CTR will be. As the novelty wears off and the law of shitty click-throughs kicks in, I would expect to see clicks reduce further,” he noted. Brand Queries See CTR Lift Meanwhile, in a study of 700,000 keywords, performance agency Amsive found keywords that triggered an Overview had an average click-through decline of 15.49%. The study also noted that while only a small percentage (4.79%) of branded keywords generated an Overview, those that did had an average click-through rate increase of 18.68%. By contrast, non-branded keywords that generated an Overview had an average click-through rate drop of 19.98%. “It’s no surprise that branded keywords are getting more clicks than non-branded keywords,” said Greg Sterling, co-founder of Near Media, a market research firm in San Francisco. “The vast majority of AI Overviews trigger when the user does an ‘informational’ search — people looking for general information,” he told TechNewsWorld. “When somebody uses a branded keyword, there’s a higher degree of intent and thus a higher CTR. There’s nothing new in this.” Ben James, founder of 404, Bittensor Subnet 17, an online 3D content creation company, explained that non-branded keywords typically drive discovery and surface diverse viewpoints. “If AI Overviews are disproportionately reducing CTR for those terms, it reinforces concerns that Overviews consolidate traffic around known brands or Google’s own properties — shrinking the opportunity space for independent publishers and startups,” he told TechNewsWorld. However, the researchers’ findings seemed nonintuitive to JD Harriman, a partner at the Foundation Law Group in Burbank, Calif. “It is not clear what searchers are doing when they see an AI Overview of a branded word,” he told TechNewsWorld. “It is likely that the searcher wanted to get to the branded site in the first place, and goes ahead and clicks through, possibly ignoring the content of the Overview and just getting where they wanted to be anyway.” Fewer but Higher Quality Clicks? “There have been several studies done by different companies that show the same thing, so this was not a surprise to me at all,” said Chris Ferris, senior vice president of digital strategy at Pierpont Communications, a public relations agency in Houston. “When AI overviews are present on search engine results pages, the click-through rate for the organic stuff falls anywhere from 35% to 70%,” he told TechNewsWorld. “This makes sense because Google has been increasingly stuffing stuff at the top of their results pages, which is pushing down the organic results, which means fewer people see them, which is depressing the click-through rate,” he explained. While click-through rates might decline on pages with Overviews, Google contends that the quality of the clicks on the page improves. “We see the clicks are of higher quality because they’re not clicking on a web page, realizing it wasn’t what they want and immediately bailing. So, they spend more time on those sites,” Head of Google Search Elizabeth Reid told the Financial Times in an interview published April 14. “The studies don’t directly disprove Google’s claim about ‘high-quality clicks,’ but they show that AI Overviews likely reduce visibility and traffic for many sites, especially on non-branded, informational queries,” said Danny Goodwin, editorial director of Search Engine Land & SMX, a digital marketing and advertising technology publication. Goodwin pointed out some deficiencies in the studies. “Neither study examined whether the pages cited in AI Overviews got more clicks than they would have in a traditional search listing,” he told TechNewsWorld. He added that Google’s concept of “high-quality clicks” is vague and unverifiable outside of Google itself. A Google spokesperson was not immediately available to comment on this story. Antitrust Red Flags For some tech watchers, the studies’ results reinforce the notion that Overviews are harmful to quality content on the web. “By answering specific queries within the search interface, AI Overviews reduce the incentive to click through to original sources, diminishing traffic and weakening the feedback loop that sustains quality content creation,” said Mark N. Vena, president and principal analyst at SmartTech Research in Las Vegas. “If original content creators lose traffic, and I do fear this, monetization becomes harder, which may lower the quantity and quality of web content,” he told TechNewsWorld. “Long-term, this could degrade the richness of information available on the open web and increase dependence on platform-controlled summaries.” Rob Enderle, president and principal analyst with the Enderle Group, an advisory services firm in Bend, Ore., added that it’s intuitively obvious why Overviews will reduce the quality of web content. “Few people who buy CliffNotes read the source material because the notes fulfill the core needs of getting the gist of the content,” he told TechNewsWorld. “In reports, most only read the executive summary, as well. If you don’t need to review the full detail, why bother unless you have a unique need for a deeper understanding?” Study findings on the impact of Overviews on website traffic could play a role in antitrust actions against Google. “Google has been under scrutiny for a while now for self-preferencing its own products and services in search,” Goodwin said. “AI Overviews are another big and bold example of Google preferencing itself, many times at the expense of content creators.” “I, and many others, would love to see actual data about how much traffic Google is actually taking for itself via AI Overviews compared to the open web,” he noted. Sterling agreed. “I do think that AI Overviews will have an impact on antitrust considerations,” he said. “AI Overviews can be seen as a form of Google self-preferencing, which will create problems for Google in Europe under the Digital Markets Act. In the U.S., the decline of clicks will reinforce perceptions of Google as a monopolist hoarding traffic.” AI Overviews Reflect Market Pressure However, Jennifer Huddleston, a technology policy research fellow at the Cato Institute, a Washington, D.C. think tank, argued that companies, including Google, continue to innovate to respond to changing consumer expectations and demands, including better ways to display search content such as AI Overviews. “The use of AI Overviews illustrates how AI is changing the nature of search, and consumers may be increasingly expecting more generative AI-type results to their queries,” she told TechNewsWorld. “While the court rejected the emergence of AI as changing the market in this case,” she noted, “the continued improvements of the product indicate how the company must respond to pressures from other market leaders and consumers, a behavior that would not be necessary in a true monopoly situation.” John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. Related Stories More by John P. Mello Jr. view all More in Search Tech

AI, Layoffs Fuel Surge in Job Scams By John P. Mello Jr. April 22, 2025 5:00 AM PT ADVERTISEMENT Proven Customer Acquisition Strategies for SMB Software Firms Discover proven tactics for scaling products, winning customers, and staying ahead. Get your free copy of the SMB Software Playbook for Expansion & Growth. Download Today. The times couldn’t be better for job scammers. Layoffs in the tech and public sectors, coupled with the malicious use of artificial intelligence, are fueling strong growth in fraud aimed at job seekers. Scams aimed at job seekers can come in a variety of forms, among them: Hire and Fire Scams — where a job seeker is quickly hired, works, and is fired before collecting a paycheck. Fake Recruiters — where a fraudster harvests sensitive information from a job seeker as part of the application process. Pay-to-Work Offers — where job seekers are told they’ve been hired, but they have to first make some upfront payments for training or equipment. “Asking for money or asking you to purchase something with the promise of a reimbursement is a scam that has been around for a long time,” said Julia Toothacre, chief career strategist at Resume Templates. “These are companies that appear legitimate and sometimes even have a website, but their purpose is to scam people out of money,” she told TechNewsWorld. “Job seekers need to be on their toes when it comes to scams,” she added. “If it feels too good to be true, it probably is.” Robust Scam Economy Growth in job scams isn’t surprising since scam activity has been generally growing globally. “The scam economy is blowing up. In some parts of the world, it’s up 3000% year-over-year,” declared Joshua McKenty, CEO and co-founder of Polyguard, provider of a solution for defending against deepfakes and AI fraud, in New York City. Job seekers in volatile labor markets will attract scammers, he continued. “When we’ve seen big waves of layoffs in tech, especially, then scamming for workers becomes really attractive because there are lots of job seekers to target,” he told TechNewsWorld. “The same is true when hiring starts to pick up again,” he said. “A lot of hiring means a lot of targets.” He explained that AI is also contributing to job scam growth. “All of the scam economy is powered by AI right now — everything from effortless translation to scam texts, emails, and job postings that all look legitimate.” AI helps scammers pass the “smell test,” added Joshua Planos, vice president of marketing at the Better Business Bureau in Lincoln, Neb. “For a scammer, that is the ultimate goal — to give off the impression that what you are providing is real,” he told TechNewsWorld. “It stands to reason that with how much AI and chatbots are being used for everything from marketing materials to internal documents, it would have an impact on employment or job market scams,” he said. Why Job Seekers Are Prime Scam Targets Scam growth is also being driven by vulnerable job seekers, added Erich Kron, a security awareness advocate at KnowBe4, a security awareness training provider in Clearwater, Fla. “An unstable world economy has created a level of concern for many people, especially when it comes to their employment and personal finances,” he told TechNewsWorld. “This uncertainty has people who would otherwise not be interested in other employment considering their options,” he continued. “Bad actors know this, and they leverage that fear and uncertainty to meet their goals.” “A fragile labor market and desperation have created the perfect storm,” added Sam Wright, head of partnerships and operations at Huntr, an online AI-powered resume builder and job search platform. “Many job seekers apply to dozens of listings a week with little feedback, so when something promising appears, they jump,” he told TechNewsWorld. AI can contribute to that fear and uncertainty, too. “Many individuals have concerns about AI replacing their roles, so they are more open to possible career changes than in the past,” Kron said. “This concern about career obsolescence is also a tool exploited by bad actors.” New Workers Targeted by AI-Powered Scams The ease of being able to post on multiple platforms, the ability to use AI to create legitimate-looking job posts, the growth in remote work, and the sensitive data shared in job scams have all fueled their growth, noted Mona Terry, chief operating officer and head of victim services at the Identity Theft Resource Center, in San Diego, a nonprofit organization devoted to minimizing risk and mitigating the impact of identity compromise and crime. Scammers are also taking advantage of job-seeking newbies. “Many individuals are looking for remote work, and for those who have not worked remotely before, or those new to the job market after having been employed for a number of years, the interview process may be new,” Terry told TechNewsWorld. “Asking for interviews on nontraditional platforms is a red flag for those in the know, but not for the average person,” she explained. “Scammers ask for information that is traditionally asked for after hiring as proof of eligibility to work — social security numbers, driver’s license, bank account numbers for direct deposit, etc. — before hiring.” Dean Boerner, a data scientist with Revelio Labs, a workforce intelligence company in New York City, noted that one reason fake job postings and phishing emails work is the vast and growing number of job postings on the internet and job seekers’ inability to sufficiently vet all those that might be of interest to them. “Just to give you an idea of the scale here, at any given time, hundreds of millions of open job postings are represented in our data, and these vary in terms of their likelihood of leading to an actual hire,” he told TechNewsWorld. “While not really a scam, the prevalence of ghost job postings contributes to the apparent multitude of opportunities, and these have proliferated as economic uncertainty has grown,” he said. “For instance, we’ve seen the ratio of hires per job postings fall from about 0.75 to below 0.5 in recent years. Part of this is also from a shortage of talent employers deem as truly qualified, and some of it may be from companies wanting to present as growing without actually growing.” How To Spot and Avoid Employment Scams For job seekers who wish to protect themselves from scammers, Diana Rothfuss, global solutions strategy director for risk, fraud and compliance at SAS, an analytics and artificial intelligence software company, in Cary, N.C. recommends that promising positions posted on a job board be cross-referenced on the company’s website to vet their legitimacy. “Only apply for jobs via the company’s website,” she told TechNewsWorld. “Very often, the scammer’s end goal is to collect sensitive personal information that can be used to commit other frauds, including identity theft.” “Don’t click the link in the job posting,” she warned. “Open the company’s website in another browser to ensure you haven’t been directed to a phishing site.” Beware of urgency, cautioned John Wilson, a senior fellow for threat research at Fortra, a cybersecurity services company in Eden Prairie, Minn. “While some jobs may be urgently needing a candidate, no legitimate company will hire anyone who has not been background checked — they can get in legal trouble — and this takes time,” he told TechNewsWorld. “Whenever you feel pressured to act quickly, slow down and check things out,” he said. “Urgency is a known persuasion technique used to influence people.” The best way to avoid these scams is to be cautious with unsolicited job offers, advised Paul Farnsworth, president of Dice, a career website for technology and engineering professionals in Centennial, Colo. “Always research the company thoroughly, be wary of any job that asks you to pay a fee upfront, and steer clear of any offers or setups that seem too good to be true,” he told TechNewsWorld. “Don’t forget the classic phishing red flags: double-check inconsistencies in email address domains, don’t share sensitive personal info, and look out for anything suspicious.” “With AI getting smarter, an uncertain economy, and the fact that these scams are low-risk and potentially high-reward for the people running them, chances are they’ll perpetuate and grow,” he said. John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. Related Stories More by John P. Mello Jr. view all More in Cybersecurity

Over the last decade, AMD has undergone a remarkable transformation. From a company once struggling for relevance, it has risen under CEO Lisa Su’s steady and strategic leadership to become a serious contender across the data center, client computing, and now embedded and adaptive edge markets. One of AMD’s fastest-rising stars is its embedded business, now armed with a broad portfolio and a sharp focus on AI. As competitors like Intel falter, AMD’s differentiated approach could put it on a trajectory to gain significant market share, particularly at the embedded edge. AMD’s Embedded Rebound and Push Into Edge AI AMD’s acquisition of Xilinx laid the foundation for its embedded success. This purchase brought a robust adaptive computing portfolio — FPGAs, SoCs, and RF technologies — that AMD has tightly integrated with its x86 CPUs, GPUs, and NPUs. In an invite-only fireside chat with analysts last week, Salil Raje, SVP and GM of AMD’s Adaptive and Embedded Computing Group, the depth of that integration was on full display. Raje outlined AMD’s five-pillar strategy: Fortifying its adaptive portfolio Improving usability for developers Growing x86 embedded market share Winning high-value custom silicon deals Leading in embedded AI AMD isn’t just positioning itself as a component vendor — it’s becoming a “platform enabler” across industries like automotive, aerospace, communications, and robotics. AMD’s Edge Over Intel in Embedded Strategy It’s clear that AMD is not playing catch-up — it’s pushing ahead where others stall. AMD has achieved revenue leadership in adaptive computing, leaving Intel’s Altera (soon to be spun out again) in its wake. In embedded CPUs, AMD holds just a 7%-8% market share but sees that as an opportunity, not a weakness. Raje said, “We believe we can grow significantly faster in that business over the next four to five years.” What makes AMD’s approach stand out? Flexibility and openness. AMD’s edge strategy doesn’t depend on any single compute architecture. Instead, it uses a modular mix of x86, Arm, GPU, and FPGA — whatever the application demands. The company also avoids a black-box approach to AI software stacks, instead partnering with ecosystem players to keep platforms open and customizable. This open strategy contrasts sharply with more closed approaches from some of its competitors, particularly in automotive and robotics. AI at the Edge: AMD’s Next Big Move Perhaps the most exciting element of AMD’s embedded strategy is its continued aggressive push into edge AI. “There will be a ChatGPT moment at the edge,” said Raje, and AMD wants to be ready for it. AMD is integrating NPUs into nearly every product, from AI PCs to embedded SoCs. The goal is simple: deliver low-latency, power-efficient AI acceleration in markets like industrial automation, medical imaging, and autonomous vehicles. AMD’s recent product launches reflect this vision. From the versatile Versal AI Edge Gen 2 — which fuses ARM cores, FPGA fabric, ISPs, and NPUs — to the powerful EPYC Turing 9005 with 192 Zen 5 cores, the company is scaling across performance tiers and verticals. It’s already winning sockets in security, networking, and auto sectors. Moreover, AMD’s AI software tools enable seamless migration from cloud-trained models to edge deployments, a unique value proposition that strengthens customer loyalty. Custom Silicon: Playing Offense, Not Defense AMD’s edge isn’t just in off-the-shelf products. Its custom silicon business, once limited to gaming consoles, is gaining traction across automotive, defense, and data center sectors. Importantly, AMD doesn’t pursue custom silicon unless it brings differentiated IP or platform value, such as integrating x86, GPU, or RF IP into unique packages. It’s a targeted, value-driven strategy that avoids commoditization. Chiplets add another layer of flexibility. AMD’s leadership in chiplet architectures allows it to offer semi-custom solutions more cost-effectively, integrating customer IP on shared platforms. As chiplet adoption increases, AMD’s ability to modularize compute elements will become a powerful differentiator. Leadership That Delivers Much of AMD’s rise can be traced back to the discipline and strategic clarity of CEO Lisa Su. Her turnaround of the company isn’t a product of bold promises but of measured execution that prioritizes innovation, product roadmaps, and market focus. That same DNA is evident in AMD’s embedded and adaptive computing. Su’s leadership enabled AMD to avoid the pitfalls that have plagued Intel — missed process nodes, delays in AI strategy, and an overdependence on legacy business lines. AMD, by contrast, now delivers products that are often competitive but ahead of the curve, particularly in power-performance ratios and time-to-market in AI. The Intel Factor: A Window of Opportunity Intel’s struggles in recent years have opened the door for AMD. From manufacturing delays to uncertainty around the Altera spinout, Intel’s position in the embedded market is vulnerable. While Intel still dominates in x86 embedded CPUs, its distracted execution allows AMD to grab market share, especially with AI reshaping workloads at the edge. AMD’s strength in heterogenous computing, openness to Arm, and software-developer focus make it far more agile in responding to evolving edge AI needs than Intel’s more rigid portfolio. If AMD executes on its roadmap and continues to differentiate with a modular platform approach, it could become the go-to supplier for a broad swath of edge workloads. A Platform for the Future AMD’s embedded business is no longer just a side bet. It’s rapidly becoming a cornerstone of the company’s long-term growth strategy. Once viewed as a niche, the embedded market is now a critical front in the broader computing landscape, especially as AI workloads migrate from centralized data centers to distributed, real-time environments at the edge. Under Lisa Su’s direction, AMD’s leadership team has positioned the company to capitalize on this shift with discipline, clarity, and a sharp focus on execution. The strategy is not just about having a broad portfolio but about how it works together. AMD is offering customers a consistent, scalable compute platform from cloud to edge, blending the flexibility of adaptive hardware with the performance of CPUs, GPUs, and NPUs. This point is especially relevant in today’s fragmented edge environments, where power efficiency, latency, and customization define competitive advantage. AMD’s modular approach, enabled by chiplets and customizable silicon, ensures customers can get precisely what they need without compromise. AMD’s Opportunity To Redefine Embedded Computing The company’s solid stance on open software ecosystems resonates in markets that have grown weary of closed, proprietary solutions. This customer-centric approach and a differentiated product roadmap make AMD more than a component supplier: it’s becoming a strategic partner across industries. As Intel contends with a new CEO, internal restructuring, and struggles to execute cleanly in the embedded and AI domains, AMD has a rare and meaningful opportunity to gain market share and mindshare. Already, the momentum is evident: new design wins, expanding market share in adaptive and embedded CPUs, and deepening traction in custom silicon engagements. The path ahead won’t be without challenges. Arm-based players, vertical integration trends, and software complexity will continue to test all major players, but AMD appears better prepared and positioned than ever. It’s clear that AMD is not just catching up; it’s redefining the rules of the embedded game. If it continues to execute with the same precision that defines its turnaround story, AMD won’t just lead in AI at the edge but will help determine what the edge becomes. What a remarkable juxtaposition AMD’s position in the semiconductor industry is today compared to just 10 years ago.

While tariffs create a clear incentive to bring manufacturing back to the U.S., this is far from an easy task. Building up the kind of manufacturing capacity the U.S. once had will take decades, as it did in China. This week, we’ll examine why it is so difficult to reverse a trend that dates back to the breakup of RCA in the 1940s — marking manufacturing as a low government priority — an issue that can’t be fixed in a few years. We’ll close with my Product of the Week, a hydrogen-powered horse robot (in prototype form) from Kawasaki named Corleo that should be far better than a motorcycle if it makes it to market. Reshoring US Manufacturing Is Difficult The call to “bring manufacturing back” resonates deeply in the American psyche, fueled by nostalgia for a perceived golden era, concerns about national security, and anxieties over global supply chain vulnerabilities. Electronics manufacturing, which encompasses everything from intricate semiconductors to consumer gadgets, is at the heart of this discussion. Yet, despite the political rhetoric and popular support, reshoring complex electronics manufacturing to the U.S. borders on impossible in the near term. The reasons are multifaceted and deeply entrenched, stemming from decades of global economic shifts, strategic investments by competitors, and fundamental differences in infrastructure, workforce, and policy environments. Why China Still Leads Electronics Manufacturing China methodically built itself into the world’s workshop for decades, particularly for electronics. This dominance wasn’t accidental; it resulted from strategic planning and the cultivation of several key advantages that the U.S. currently cannot match. Labor Dynamics and Scale While often cited, the simple “cheap labor” argument is now oversimplified, though it was the initial catalyst. Chinese labor costs have risen significantly. However, the sheer scale of the available workforce, coupled with decades of experience, specifically in electronics assembly and production processes, creates efficiency and flexibility that is difficult to replicate. More importantly, this vast labor pool allowed for the creation of manufacturing ecosystems on a scale unseen elsewhere. Foxconn City in Shenzhen, employing hundreds of thousands of workers in a single integrated campus, exemplifies this scale. It’s a model with no parallel in the U.S. Even as automation increases, the existing skilled and semi-skilled labor base provides a foundation and adaptability critical for the rapid product cycles common in electronics. Purpose-Built Infrastructure China didn’t just attract factories; it built the entire logistical ecosystem to support them. This infrastructure includes massive, modern ports optimized for container shipping, extensive high-speed rail networks for domestic component transport, dedicated manufacturing zones with reliable utilities, and dense clusters of suppliers. Component manufacturers, mold makers, testing facilities, and assembly plants often exist in close geographic proximity, enabling rapid iteration, troubleshooting, and production scaling — a robust supply chain ecosystem. While extensive, the U.S. infrastructure is geared mainly toward consumer logistics and lacks the hyper-specialized, manufacturing-centric density found in regions like the Pearl River Delta or the Yangtze River Delta. Recreating such an integrated infrastructure in the U.S. would require decades and astronomical investment. Robotics and Automation Adoption Counterintuitively, China is also a leader in manufacturing automation. Facing rising labor costs and striving for higher precision, Chinese manufacturers, often aided by government initiatives, have invested heavily in robotics and advanced manufacturing technologies, rapidly increasing their robot density. They possess the scale for manual assembly when needed and increasingly sophisticated automated lines for high-volume, high-precision tasks. This combination of flexible manual labor capacity and advanced high-precision automation enables Chinese firms to stay competitive across a wide range of manufacturing complexities. While the U.S. has advanced automation capabilities, the scale of deployment within integrated electronics ecosystems lags behind China. Aggressive Government Financial Support The Chinese government has treated the development of its manufacturing sector, especially in strategic areas like electronics, as a national priority. To support this goal, it has provided massive, sustained financial support through various mechanisms: direct subsidies to manufacturers, low-interest loans from state-owned banks, tax incentives, government funding for research and development, land grants, and, sometimes, less tangible benefits like expedited regulatory approvals. This coordinated, long-term state backing created an environment where companies could invest heavily in capacity and technology with reduced financial risk, fostering the sector’s rapid growth. Organizations like the IMF have analyzed this approach as having significant trade implications. Decades-Long Hollowing Out of U.S. Manufacturing The rise of China’s manufacturing prowess coincided with a decades-long decline in U.S. manufacturing capacity, particularly in electronics assembly and component production — excluding some high-end areas like advanced semiconductor design. This wasn’t a sudden collapse but a slow bleed driven by corporations seeking lower production costs and focusing on design, marketing, and software — the higher-margin ends of the value chain. The consequences are profound. Factories closed, equipment was sold off or scrapped, and crucially, entire generations of skilled manufacturing workers retired or moved into other sectors. The intricate networks of specialized suppliers that support complex manufacturing dwindled. Institutional knowledge — the practical expertise in running complex production lines, managing supply chains, and training technicians eroded. Rebuilding this isn’t just about constructing buildings; it’s about resurrecting an entire industrial ecosystem, complete with its human capital and specialized knowledge base. As analysis suggests, millions of jobs were lost over decades. If it took decades to dismantle, it stands to reason that it will take decades to rebuild, assuming the will and resources are consistently available. Immense Hurdles of Rebuilding: Time, Investment, and Policy Even if the U.S. were to embark on a serious reshoring effort, the practical challenges are staggering. Building Takes Time Constructing a large-scale, state-of-the-art manufacturing facility, like a modern semiconductor fabrication plant (fab), is a multi-year endeavor. It involves site selection, complex permitting processes (environmental, zoning), massive construction, installation of highly specialized equipment, and rigorous calibration and testing. A single fab can take three to five years or more, from groundbreaking to full production, costing upwards of $15-$20 billion, assuming no significant delays. To make a dent in reshoring electronics, the U.S. would need not just one but dozens of such large facilities, plus countless smaller suppliers — a monumental undertaking spanning decades. Need for Long-Term, Unshakeable Incentives Building these facilities requires colossal capital investment. Companies undertaking such investments need certainty regarding the long-term economic environment to ensure a return on investment. Incentives like tax breaks or tariff relief are often discussed. However, U.S. policy, particularly regarding trade and tariffs, has proven highly volatile, creating uncertainty that disrupts predictability and discourages investment. Tariffs implemented by one administration can be altered or removed by the next or even adjusted month-to-month based on geopolitical shifts or trade negotiations. For a factory with a 10–20-year payback period, the risk that crucial incentives could disappear within a single presidential term is a major deterrent. Investors require policy stability that extends far beyond typical election cycles — something historically difficult to guarantee in the U.S. system. Without credible, long-term, and perhaps even legislatively locked-in incentives, the financial risks of massive reshoring investments are often too high for private companies to bear alone. US Lacks Skilled Labor for Electronics Work Beyond physical infrastructure and policy, there’s the human element. Skills Gap Decades of de-emphasis on manufacturing have led to a significant mismatch between the skills needed for advanced electronics manufacturing and those prevalent in the U.S. workforce. Modern electronics production requires highly skilled technicians proficient in robotics, process control, cleanroom protocols, quality assurance, specialized equipment maintenance, and engineers with deep expertise in manufacturing processes. While the U.S. excels in design and engineering, the pipeline for skilled manufacturing technicians and engineers has shrunk considerably. In contrast, China produces vastly more engineers and technicians annually, many specifically trained for manufacturing roles, although challenges remain in aligning skills with industry needs globally. Rebuilding America’s vocational and technical training systems to meet this demand will require long-term investment and a shift in educational and cultural priorities. Workforce Discrepancy There’s a stark contrast between public sentiment and individual career choices. Surveys consistently show overwhelming public support (around 80%) for bringing manufacturing back to the U.S., a sentiment that holds across party lines and contexts. Yet only 20%–25% say they would consider working in a factory. This disconnect reflects changing career aspirations, perceptions (sometimes outdated, though improving according to some studies) of factory work as monotonous or physically demanding, and the availability of jobs in the service, tech, and gig economies. Even with training programs and improving perceptions, attracting and retaining a sufficient number of workers willing to take on demanding manufacturing roles, potentially at wages that must still compete globally (even with automation), presents a significant social and economic challenge. Catching Up: The Price Tag of Reshoring China’s manufacturing dominance was built on decades of massive, state-directed investment, which created systemic advantages and sometimes led to global overcapacity issues. For the U.S. to catch up, particularly in a mature industry where China already has established economies of scale and integrated ecosystems, the required investment would likely need to be even larger and more sustained. Closing that gap isn’t just about subsidizing factory construction. While initiatives like the CHIPS Act represent important progress, they address only a portion of the broader manufacturing ecosystem. Achieving true competitiveness would require coordinated public and private investment in workforce development, infrastructure, supply chains, and advanced manufacturing capabilities. Sustaining such an effort over decades would demand a level of political consensus and fiscal commitment that is historically difficult to maintain in the U.S. Wrapping Up The desire to reshore electronics manufacturing is understandable and driven by legitimate economic and security concerns. However, the practical realities paint a picture of immense difficulty, particularly in the near term. China’s entrenched advantages in scale, infrastructure, government support, and, increasingly, automation create a formidable barrier. Decades of U.S. manufacturing decline have left deep gaps in physical capacity, supply chains, and skilled human capital. Rebuilding requires not only time — years to build individual plants and decades to reconstruct ecosystems and workforces — but also unprecedented levels of sustained investment and, critically, long-term policy stability that is antithetical to recent U.S. political history. While targeted investments in strategic niches like advanced semiconductors are possible and underway, the wholesale return of the broad electronics manufacturing ecosystem seen in Asia remains, for the foreseeable future, an aspiration facing near-insurmountable obstacles. It’s a multi-decade marathon requiring unwavering national commitment, not a sprint that can be won with short-term policies or wishful thinking. Corleo: Kawasaki’s Hydrogen-Powered Horse Robot Imagine a future where transportation isn’t limited to wheels or wings, but steps confidently over rough terrain. Kawasaki is bringing that vision closer with “Corleo,” a fascinating hydrogen-powered, rideable quadruped robot — essentially, a robotic horse unveiled for Expo 2025 Osaka. This isn’t just science fiction — it’s a glimpse into practical future applications, as seen in Kawasaki’s concept video: Why a robot horse? Think about the advantages. Real horses are incredibly sure-footed and capable of navigating landscapes that would stop most motorcycles or ATVs cold. However, they require significant upkeep — feeding, stabling, and veterinary care — while motorcycles lack that all-terrain agility. Corleo aims for the best of both worlds: the potential sure-footedness of a quadruped — similar to advancements seen in other four-legged robots — combined with the lower maintenance of a machine powered cleanly by a 150cc hydrogen engine that generates electricity for its legs. The potential applications are exciting. Corleo could be invaluable for remote rescue operations. Picture sending a team of these into disaster areas or rugged wilderness where vehicles can’t reach. If developed with autonomous capabilities, Corleo could even venture into dangerous zones alone to locate and extract individuals, minimizing risk to human rescuers. Future Possibilities for Kawasaki’s Robot Horse Beyond rescue, imagine other uses: military transport in challenging environments (effectively bringing back the cavalry, but robotic!), unique, eco-friendly tour vehicles offering silent traversal through nature parks, or even as a fascinating, low-maintenance alternative for someone wanting an exotic pet without the biological demands. What could Corleo evolve into? Perhaps faster, more agile versions, integrated with advanced AI for complex tasks or specialized models for specific industries. It represents a stepping stone toward more versatile and adaptable robotic mobility, though the current capabilities are limited compared to the concept vision. Honestly, forget the sports car — a hydrogen-powered robot horse that can traverse mountains? Now that would make an incredible birthday gift for someone adventurous (hint: my birthday is coming up). Kawasaki’s Corleo isn’t just a robot; it’s a four-legged stride into the future — and my Product of the Week.

As AI continues reshaping the classroom, a new multinational study reveals that students are more worried about its impact than teachers or administrators — raising fresh concerns about critical thinking, misuse, and the future of learning. The post AI Raises Bigger Concerns for Students Than Teachers, Admins: Study appeared first on TechNewsWorld.

NTT is one of the most powerful technology conglomerates in the world. As a telecommunications giant, the company operates a global network infrastructure, ranks among the top three data center providers worldwide, and employs more than 330,000 people. Through its NTT Data subsidiary, it boasts deep expertise in cybersecurity, next-gen networking, quantum research, and enterprise IT services. With these assets, NTT should be positioned as an industry-defining force in the AI-powered future. However, at its recent NTT Research Upgrade 2025 event, it chose not to lead. Instead, a bold, clear articulation of vision and ambition became a disjointed series of ideas, framed more as academic explorations than strategic imperatives. Lofty language, vague metaphors, and philosophical asides dominated the company’s core messaging. While the technology showcased was real and impressive, the leadership tone was tentative — more deferential to partners than a directive to the market. In short, NTT missed the moment, which is a strategic misstep for a company with its capabilities. Ambitious Language, Unclear Strategy NTT Executive Chairman Jun Sawada opened the keynote with expansive, almost poetic language. He spoke about “upgrading reality,” building a “network of AIs,” and promoting a “pluralistic value society.” He referenced philosophical inquiries and the creation of the Kyoto Institute of Philosophy, aiming to bring together Eastern and Western thought in shaping the future of human-AI coexistence. These are intriguing themes. However, they lacked anchoring in a keynote intended to showcase technology leadership — which turned into a couple of fireside-style chats. There was little connection to product roadmaps, customer impact, or near-term business outcomes. Sawada spoke of IOWN — the company’s ambitious, Innovative Optical and Wireless Network — but left unexplained how this infrastructure uniquely enables or differentiates NTT in the AI race. Sawada mentioned the importance of connecting AIs to build a “heterogeneous world” but stopped short of detailing what NTT is building to realize that goal. The result? The keynote felt more like a symposium on the philosophy of technology than a declaration of leadership. For NTT, whose customer base includes some of the world’s most demanding enterprise and government clients, that was a missed opportunity to inspire confidence. Research Focus Felt Like Strategic Retreat Kazu Gomi, CEO of NTT Research, doubled down on the company’s commitment to basic research. He described NTT’s Silicon Valley-based research labs and their work in applied physics, cryptography, and cardiovascular bio digital twins. To his credit, he was refreshingly candid in admitting that much of this research may not become a commercial product in the near term. Again, this level of scientific humility is understandable from a university lab. However, from a Fortune Global 100 enterprise, it comes across as strategic deferral. Gomi introduced a new concept called “Physics of AI,” which is meant to demystify how AI systems work and increase their trustworthiness. However, his description leaned heavily on metaphors, such as apples falling from trees and Newtonian physics, and lacked tangible proof points or technical framing. Yet, there was no discussion of frameworks, metrics, or deliverables. There is no sense of how this new initiative would generate a competitive advantage for NTT or its customers. There was no call to action for industry partners; it was just an abstract pitch. In a world where competitors like Microsoft, Nvidia, Amazon, and Google are integrating AI into everything from data centers to developer tools, NTT’s insistence on staying in the research phase feels like a reluctance to lead. Snowflake and NTT Data Filled the Vacuum Ironically, the most substantive content of the event came from outside NTT’s core corporate leadership. Snowflake CEO Sridhar Ramaswamy delivered a practical, compelling roadmap for enterprise AI adoption. He described how Snowflake helps clients extract value from their internal data, build AI copilots, and deploy production-grade chatbots. He emphasized trust, efficiency, and ease of deployment—core values that enterprise buyers care deeply about. NTT Data CEO Abhijit Dubey provided a similarly focused perspective. He highlighted four innovation areas: quantum computing, voice mining, attribute-based encryption, and trusted data spaces. Each was tied to real-world customer stories: BMW, Japan Post Bank, and smart city pilots. These were clear, testable use cases tied to value delivery. But therein lies the disconnect. These leaders made the case for innovation. NTT itself did not. While Snowflake and NTT Data sounded like organizations building and delivering the future, NTT Corporation remained in the background — host to a conversation it should have been driving. Failure To Connect the Dots NTT has everything it needs to be a global technology leader in the AI era: A top-tier worldwide fiber and wireless network Massive data center presence, especially in high-growth markets like India Deep talent in quantum science, cybersecurity, and encryption An enterprise customer base that spans industries and continents An IT services division (NTT Data) that is among the largest in the world But Upgrade 2025 failed to tie these assets into a coherent strategic narrative. There was no aggressive positioning of IOWN as the backbone for AI-native networking. There was no clear articulation of how NTT’s data center footprint enables sovereign AI or edge intelligence. No insight was provided into how NTT’s research labs fuel product innovation pipelines, and perhaps most surprisingly, there was no bold statement about how NTT intends to compete in an industry that is moving at unprecedented speed. NTT didn’t use its stage to declare leadership. It used it to float ideas and host guests. For a company of its size, that’s a passive stance and a missed opportunity to drive the conversation forward. The Stakes Are Higher Than Ever Enterprise buyers are not looking for abstract promises. They want trusted partners who can deliver secure, scalable AI infrastructure, integrate with their systems, and help them navigate complexity. Government clients want solutions they can control and deploy locally. Telecom and cloud operators are battling to redefine the role of connectivity in an AI-native world. NTT could lead all of these markets. It has the assets. It has the talent, and, most crucially, it has the credibility. But to compete, it must first lead with confidence and clarity. Upgrade 2025 was a chance to assert its leadership and showcase NTT’s technology, purpose, business model, and vision for where the AI-powered enterprise is going. Instead, it chose the safety of abstraction. Final Thoughts: A Stage Missed NTT doesn’t need consumer recognition. It doesn’t sell smartphones or search engines. Its impact is felt behind the scenes — where networks are built, data is stored, and enterprise systems are connected. That makes forums like Upgrade 2025 especially important. They are the rare public moments where NTT can show the market what it stands for. Unfortunately, the company didn’t do itself any favors at this year’s event. It did not strengthen its position as an unabashed thought leader. It did not clarify how its formidable portfolio of technologies, across infrastructure, research, and services, comes together to solve enterprise and societal challenges — and it certainly did not project urgency. In a market where AI adoption is now a race — not a roadmap — NTT’s hesitance to lead was palpable. The company may have world-class technology and ideas, but at Upgrade 2025, it left leadership on the table.

Western automotive giants face an existential challenge as China’s carmakers surge ahead. Much of China’s edge comes from how its automotive sector is structured: streamlined, fast, and incredibly tech-forward. R&D cycles for Chinese electric vehicles (EVs) typically span just nine to 18 months. In contrast, many Western automakers require five to seven years to bring a new vehicle from concept to production. If this gap isn’t closed — and fast — the Western auto industry could become irrelevant by 2030. Let’s talk about how Intel and others could save the auto industry. Then, we’ll close with my Product of the Week: the HP EliteBook Ultra G1i 14-inch AI PC. China’s Fast-Moving Auto Sector One critical reason for China’s lead is its organizational agility. Chinese car companies operate under more linear leadership models, where decision-making is centralized and fast. There’s minimal bureaucracy, and product development teams are tightly aligned with executive leadership. In contrast, Western companies often operate under legacy structures built over decades, with siloed departments and multilayered chains of approval. These structures kill speed. In today’s hypercompetitive environment, time is everything. Why Chinese Carmakers Delay Hardware Decisions Another strategic advantage Chinese carmakers have embraced is choosing hardware last. Instead of locking in chipsets, cameras, and sensors early in development, they build vehicles on flexible, software-defined architectures. This approach allows them to integrate the most advanced technologies available at the time of launch, ensuring they’re not stuck with obsolete systems by the time the car hits the road. Western automakers still operate in reverse: they make hardware decisions early, which locks the vehicle into slower and less capable tech. That’s why you can buy a brand-new car and still be disappointed to find that the infotainment system and other tech features haven’t advanced much from what was available a decade ago. Enter Intel and the Tech Sector At this point, Intel and other tech giants could dramatically alter the trajectory of Western automotive companies. Intel has deep roots in high-speed innovation and leading technology transitions. It could provide crucial expertise in agile development practices, hardware/software integration, and modular platform design. One key area is photonics, a revolutionary technology that enables data to travel as light rather than electrical signals. Intel is among the world leaders in silicon photonics and could guide carmakers in replacing expensive, heavy, and increasingly bandwidth-constrained copper wiring with fiber-optic solutions. Photonics doesn’t only reduce weight — it drastically improves data transmission speeds, making it ideal for next-generation vehicles loaded with sensors and AI modules. Intel, AMD Partnership Could Save Western Carmakers In the early days of the PC revolution, Intel and AMD drove massive innovation by functioning as second-source partners, providing interchangeable processors that helped standardize the personal computing landscape. This model could be revived for automotive computing. Currently, Qualcomm and Nvidia dominate the automotive AI chipset space, but a renewed Intel-AMD alliance could give automakers a flexible, competitive alternative. To unlock this potential, the two firms should consider standardizing common sockets or board layouts, much like they did in the x86 era. Standardization would allow automakers to adopt a single-vehicle compute architecture and choose between Intel and AMD for different trim levels to overcome product shortages or regional variants, dramatically reducing development costs and improving sourcing resilience. In addition, both AMD and Intel are highly scalable at the socket level, allowing better upgrade paths and enabling last-minute or post-sales enhancements that could generate new revenue for dealers. AI Will Reshape the Future of the Auto Industry AI is poised to radically reshape the automotive landscape. The rise of Cars-as-a-Service (CaaS) models, autonomous vehicle fleets, and real-time predictive maintenance will fundamentally change the economics of mobility. In the future, owning a car may be less attractive, especially in urban environments. Consumers will demand smarter, more connected, and constantly improving vehicles. Automakers will need to shift from building static machines to offering dynamic, software-defined platforms. Tech companies understand this realm deeply. Firms like Intel, Google, and Amazon are already experts in cloud integration, edge processing, real-time analytics, and software monetization. By partnering with these firms, car companies can accelerate their transition to the automotive-as-a-platform model. Wrapping Up: Clock Is Ticking for Western Carmakers China isn’t just ahead in the EV race — it’s accelerating. Brands like BYD, Nio, and XPeng already deliver world-class electric vehicles at unbeatable prices. They’re expanding globally and investing heavily in autonomy and AI. Meanwhile, Western carmakers remain hamstrung by legacy supply chains, outdated software stacks, and glacial product cycles. Without bold, immediate action, many could face extinction by 2030. This is where Intel and its peers offer more than chips — they offer a lifeline. By helping legacy manufacturers reimagine their cultures, processes, and architectures, tech companies can serve as strategic partners in a much-needed reinvention. The opportunity is immense, but so is the risk. Carmakers that don’t rapidly adopt fast-cycle development, modular compute architectures, photonics, and AI-driven service models will not only lose market share — they’ll become irrelevant. If Intel, AMD, and their ecosystem allies work hand-in-hand with automakers to build next-generation vehicles, they won’t just help an industry survive — they’ll help it thrive in a world where speed, intelligence, and adaptability decide who leads and who’s left behind. HP EliteBook Ultra G1i 14-inch AI PC HP’s new EliteBook Ultra G1i 14-inch is one of the first next-generation AI PCs to hit the market. It’s tailored for professionals who value security, sustainability, and premium user experience. This sleek business laptop features Intel’s Core Ultra 5 or 7 processor, delivering a whopping 48 TOPS (tera operations per second) of AI acceleration power. This AI acceleration can improve performance by up to 2.2x for demanding tasks like video editing or webinar production. Unlike gaming laptops that sacrifice battery life for graphics performance, the EliteBook Ultra G1i is built for real-world productivity. Its lightweight design, compact charger, and extended battery life are optimized for business travel and hybrid work. Despite its slim profile, it packs surprising performance muscle. The EliteBook Ultra’s showpiece is its stunning 2.8K OLED display, available in touch and non-touch options. This high-resolution screen offers vibrant colors, deep blacks, and razor-sharp detail — ideal for professionals who value visual clarity in their daily workflow. HP EliteBook Ultra G1i 14 inch Notebook Next Gen AI PC Wolf Pro Security Edition (Image Credit: HP) Security is another area where this notebook shines. It includes HP Wolf Security, a comprehensive endpoint defense suite that provides protection at every layer of the device. It’s also certified as a Microsoft Secure Core PC, which meets the strictest requirements for firmware protection, hardware-based root-of-trust, and identity safeguards, which are ideal for handling sensitive corporate data. One of the standout additions is support for the full Poly Camera Pro experience. This camera system uses AI to improve lighting, auto-frame subjects, suppress background noise, and enhance video clarity, giving users unmatched control over how they appear in virtual meetings. For leaders and executives constantly on calls, it’s a game-changer. HP goes big on sustainability. The EliteBook Ultra is one of the most eco-conscious laptops in its class, made with recycled metals and ocean-bound plastics throughout the chassis and packaging. The laptop’s unique Atmospheric Blue finish isn’t just eye-catching — it underscores HP’s effort to move away from generic enterprise design. Ultimately, this laptop is designed with the modern leader in mind. Its next-gen AI capabilities and unmatched build quality complement its strong security foundation, efficient performance, and high-end visuals. It’s not just a work tool — it’s a statement of intent for professionals who demand the best. For anyone in search of a future-proof business notebook that marries style, substance, and sustainability, the HP EliteBook Ultra G1i 14-inch Notebook Next Generation AI PC is one of the most compelling choices available today, and it’s my Product of the Week.

Collection of Private Data Makes Mobile Apps Fat Target for Hackers By John P. Mello Jr. April 9, 2025 5:00 AM PT ADVERTISEMENT Proven Customer Acquisition Strategies for SMB Software Firms Discover proven tactics for scaling products, winning customers, and staying ahead. Get your free copy of the SMB Software Playbook for Expansion & Growth. Download Today. Mobile applications are quietly attracting more and more malevolent attention — and for good reason. They contain a trove of private information about their users. In the iOS universe alone, 82.78%, or about 1.55 million apps, track private user data, according to the trends tracker Exploding Topics. Mobile apps have also proven to be particularly vulnerable attack surfaces for cybercriminals. “Invisible” points of ingress and egress inside mobile apps can be compromised before legacy security tools even detect a breach. Those points include API calls, background syncing, and push notifications. Satish Swargam, principal security consultant at Black Duck Software, an application security company in Burlington, Mass., explained that a mobile user can grant permissions in mobile apps. “Most users do not diligently apply the permissions and broadly grant permissions, allowing malicious apps to exploit these invisible points,” he told TechNewsWorld. What’s more, legacy tools often don’t identify suspicious behavior until it’s too late. AI-powered fraud can bypass multi-factor authentication, exploit memory-related bugs and vulnerabilities, and hijack transactions in real time. “AI has changed the entire landscape for protecting mobile consumers, mobile transactions, mobile revenue, and mobile experiences. It’s lowered the barrier to creating attacks,” said Tom Tovar, CEO of Appdome, maker of a security and integration platform for mobile developers and enterprise professionals, in Redwood City, Calif. “I think we’ve seen a real dark renaissance around the use of AI to create attacks, enhance them, amplify them, and levy them against more and more consumers more easily than ever,” he told TechNewsWorld. “If you’re in the defense business, it’s an amazing time,” he said. “But if you’re just an average everyday consumer, it’s a pretty scary time.” “AI-powered attacks both in the real world and with mobile apps are making it easier and faster for threat actors to compromise systems,” added Chris Hills, chief security strategist at BeyondTrust, a maker of privileged account management and vulnerability management solutions in Carlsbad, Calif. “AI trained for malicious purposes can easily scan, discover, expose, and exploit flaws much more quickly than a normal human could ever,” he told TechNewsWorld. “This is why the fight to harness AI for good purposes is so important.” Mobile App Design Lacks Built-In Security Mobile apps are tempting targets for threat actors because they’re everywhere and packed with valuable information, said T. Frank Downs, senior director of proactive services at BlueVoyant, an enterprise cybersecurity company in New York City. “Think about all the personal data your apps have access to — from your location and contacts to your financial details,” he told TechNewsWorld. “With everyone constantly glued to their phones, the potential for data harvesting is enormous. Plus, the mobile app landscape is so diverse, with lots of operating systems and app stores, making it challenging to roll out security measures that fit every scenario.” In addition, many mobile apps aren’t securely designed. “Apps leak everything attackers need without resistance,” maintained Chris Wingfield, senior vice president for innovations at 360 Privacy, a digital privacy and security services provider in Nashville, Tenn. “Mobile apps constantly emit soft identifiers such as install IDs, ad SDK metadata, and analytics payloads that expose device location and fingerprinting data,” he told TechNewsWorld. “None of it was designed for security, as it was originally designed for attribution.” “Threat actors don’t need root access,” he said. “They just need the data exhaust. And mobile applications give it to them quietly, at scale, across millions of sessions. It’s one of the most reliable recon surfaces in use today.” Tovar maintained that the security model for mobile applications is designed around regulatory compliance, not stopping fraud, account takeovers, or scams. “It’s a perfect place for attackers to spend their time,” he said. “They’re going to follow the money, and if more people are transacting on unprotected mobile applications, it’s a veritable green field for attackers,” he added. Security Gaps Leave In-App Activity Exposed Blackhat hackers are also taking advantage of many organizations’ focus on backend security at the expense of endpoint security. “Many existing schemes focus on backend analytics or user behavior signals, which don’t detect or stop threats occurring directly on the device or within the app,” explained Kern Smith, vice president of global solutions engineering at Zimperium, a mobile security company headquartered in Dallas. “This leaves gaps for malware, runtime manipulation, and credential theft,” he told TechNewsWorld. Downs acknowledged that server-side protections and analyzing user activity to catch odd behavior are crucial security measures, but added, “They often miss the mark when it comes to securing the app itself — things like app logic, data storage, and communication can still be vulnerable. This backend-heavy approach can leave some doors wide open for attackers who know how to circumvent traditional defenses.” That approach also ignores what many malicious actors are really seeking. “Most protection schemes still assume the threat is credential-based,” Wingfield said. “However, modern targeting can start before an account even exists.” “Ad SDKs, analytics tools, and attribution networks quietly collect a stream of metadata — IP-based geolocation, device model, OS version, time zone, motion events, and ad IDs,” he explained. “That telemetry leaves the app immediately — unencrypted, unaudited, and often unnoticed,” he noted. “None of it hits the backend, so traditional fraud tools don’t see it, and behavioral models don’t flag it,” he continued. “Meanwhile, that stream gets stitched together across apps to map movement, infer routines, and cluster identities by place and pattern. The gap isn’t just technical; it’s conceptual. We’ve been protecting credentials while the telemetry is what’s being harvested.” Server-Side Risks Still Dominate Mobile Threats Nevertheless, there’s a solid rationale for focusing on backend apps and APIs. “The mobile app has data for one user. The server side has data for all users,” said Jeff Williams, CTO and co-founder of Contrast Security, a runtime security company in Los Altos, Calif. “While there are some interesting risks on the client side, almost all of the critical risks are on the server side,” he told TechNewsWorld. “The risk is very asymmetric, and most of it falls on the server.” “Opportunities for direct attacks on a mobile app are pretty limited,” he added. “Generally, attackers don’t listen for connections. They reach out to servers instead.” Eric Schwake, director of cybersecurity strategy at Salt Security, an API security provider in Palo Alto, Calif., maintained there is a trend toward integrating in-app protection alongside traditional backend security measures. “This trend arises from the understanding that mobile applications are becoming increasingly susceptible to attacks that circumvent backend defenses and strike directly at the app,” he told TechNewsWorld. “In-app protection enhances security by reinforcing the app against tampering, reverse engineering, and runtime attacks,” he said. “This method is essential to tackle the changing threat landscape and defend against advanced attacks aimed directly at the app.” John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. Related Stories More by John P. Mello Jr. view all More in Mobile Apps

Gartner Identifies 12 Disruptive Technologies for Future Business Systems By John P. Mello Jr. April 8, 2025 5:00 AM PT ADVERTISEMENT SMB Software Companies: Playbook for Expansion & Growth Proven strategies to scale your company, grow market share, and win more customers. Download the free playbook today. AI will become the new UI, spending will increase to counter disinformation, and increased use of Earth intelligence gathered from space will be disruptive technologies in the near future, according to a report released Monday by global technology research firm Gartner. The three technologies were among 12 that Gartner predicted would define the future of business systems. Technology leaders must prioritize these over the next five years, as they present competitive opportunities in the near term and will eventually grow to become standard throughout businesses, Gartner contended. “Technology leaders must take action now to gain a first-mover advantage with these technologies,” Gartner Vice President Analyst Bill Ray said in a statement. “Innovative advancements like generative-AI-enabled code architecture, disinformation security, and Earth intelligence will provide the differentiation needed to help enterprises pull ahead of the pack in terms of data and product offerings.” AI Replaces Staid UI Gartner maintained that gen-AI solutions using free-form text and multimedia inputs/outputs will displace the conventional form-oriented sequential UI in established enterprise applications and enable new user scenarios. A user interface provides a way for a human to navigate and communicate with a machine naturally, explained Sandi Besen, an applied AI researcher at IBM and Neudesic, a global professional services company. “Since language models enable a system to communicate with humans natively through voice, text, or visually, we might not have as much need for a ‘user interface’ the way we know it today,” she told TechNewsWorld. “If you can interact with the computer using natural language, you don’t really need a secondary UI,” said Rob Enderle, president and principal analyst of the Enderle Group, an advisory services firm in Bend, Ore. “The AI becomes your user interface,” he told TechNewsWorld. “Much like you don’t need Windows to talk to someone else, you won’t need a UI to interact with your computer.” Besen pointed out, however, that some things are still difficult to control, like input validation, without a UI. “We are seeing the emergence of a ‘dynamic UI’ that adapts based on the inputs the user needs to provide the LLM rather than forcing the user to follow a predefined flow,” she said. Gartner Vice President Analyst Ray Valdes added in a statement, “To remain competitive, traditional enterprise application software vendors will need to refactor applications to serve composable gen-AI solutions that are invoked on demand via textual and multimodal prompts.” Because of that, Gartner predicts that by 2029, more than 50% of user interactions linked to enterprise business processes will leverage large language models to bypass the UI layer in traditional enterprise applications, up from less than 5% today. New Interest in Disinformation Security Gartner also predicts that by 2030, at least half of enterprises will have adopted products or services to address disinformation security, up from less than 5% in 2024. It explained that disinformation security is an emerging discipline focused on threats from outside the corporate-controlled network. It includes a suite of technologies, such as deepfake detection, impersonation prevention, and reputation protection, that can address disinformation to help enterprises discern trust, protect their brand, and secure their online presence. “Attackers are using AI to impersonate executives, manipulate communications, and erode confidence in what people see and hear online,” said Patrick Tiquet, vice president for security and architecture at Keeper Security, a password management and online storage company in Chicago. “For enterprises, this means protecting infrastructure isn’t enough,” he told TechNewsWorld. “They also need to secure the people behind it, ensuring digital interactions remain authentic and trusted.” Ishpreet Singh, chief information officer at Black Duck Software, an applications security company in Burlington, Mass., explained that the rapid evolution of artificial intelligence technologies, such as deepfakes, generative AI, and automated bots, enables malicious actors to create highly realistic and targeted false narratives at unprecedented scale and speed. “These sophisticated disinformation campaigns can quickly influence public perception, distort market realities, and undermine organizational credibility, directly threatening brand value and long-term stakeholder trust,” he told TechNewsWorld. Gartner Senior Director Analyst Alfredo Ramirez IV added in a statement: “Disinformation attacks use external infrastructure like social media and originate from areas with limited legal oversight. Tech leaders must add ‘disinformation-proofing’ to products by using AI/machine learning for content verification and data provenance tracking to help users discern the truth.” Earth Intelligence for Business Insights Gartner also predicts that by 2028, 80% of major Earth surface assets globally will be monitored by active satellites. Earth intelligence uses AI to analyze satellite, aerial, and ground data to monitor Earth’s assets and activities, providing insights for decision-making, it explained. “That doesn’t mean maps and charts,” Ray said. “Earth intelligence is delivering numbers on global nickel production, theme park revenue, and the health of wheat crops, to name just a few.” Earth intelligence also includes accurate weather forecasting. “Weather events can adversely impact logistics and supply chains, so being able to better predict them should allow companies to more effectively assure their operations,” Enderle noted. “In addition,” he continued, “weather events can put sites, customers, and employees at risk, so having a better predictive tool will allow companies to better protect these assets.” Given the breadth of applications, Gartner maintained that Earth intelligence is applicable to all industries and enterprises. Defense has been the first adopter, but improvements in data quality and analysis techniques have rapidly expanded the use cases. “Earth intelligence applies to every business,” Ray added. “Enterprises can gain an early advantage by creatively and strategically applying Earth intelligence to significantly enhance specific functionalities of existing systems or to compete via net new capabilities.” Disruptive Dozen Gartner organized its Disruptive Dozen into four categories: Application, Infrastructure, Data, and Security. Application disrupters include intelligent simulation, AI-composed applications, and polyfunctional robots. “Polyfunctional robots will be capable of performing an increasing number of jobs currently done by humans,” Enderle said. “Putting in place the support structure for these robots and programs to retrain or eliminate the then redundant employees will require a great deal of planning, time, and effort and will certainly disrupt operations during implementation, particularly if the site is unionized.” Infrastructure disrupters include algorithm-aligned silicon, low power/high compute, and sensor fusion. Among the data disrupters are Earth intelligence, hypersynthetic data, and domain language models. “Hypersynthetic data will make creating LLMs far easier and faster,” Enderle said. “However, the potential for this data to be unreliable increases significantly the farther you get from data with a more factual foundation. Quality has suffered significantly during the AI rollout, and hypersynthetic data could increase that exposure significantly.” Besen explained that domain language models are LLMs trained extensively within a “domain” of knowledge rather than on general data like GPT-4o. “They will be important, particularly for tasks that require a very deep understanding of specialized subjects like biology or chemistry, and which don’t require knowledge or the ability to reason outside their specific domain,” she said. “A domain model for biology might be excellent for interpreting patient symptoms and explaining a physician’s diagnosis to a user — but not helping them plan their upcoming travel itinerary.” Security disrupters include digital ethics, disinformation security, and preemptive cybersecurity. “Preemptive cybersecurity stands out because it shifts security from reaction to prediction,” Tiquet explained. “As threat actors increasingly weaponize AI and automation, waiting for an alert is no longer enough,” he continued. “Preemptive systems use AI-driven insights to uncover vulnerabilities, misconfigurations, and anomalies before they’re exploited.” “It’s a disruptive change because it forces organizations to rethink traditional tools and adopt continuous, intelligent threat anticipation across every layer — from endpoints to identities to cloud workloads,” he said. Singh added, “This forward-looking approach is especially critical in today’s ever-evolving digital landscape, where the increasing complexity of interconnected systems — including IoT devices, cloud services, and remote work platforms — creates a broader and more intricate array of security challenges.” John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John. Leave a Comment Click here to cancel reply. Please sign in to post or reply to a comment. New users create a free account. Related Stories More by John P. Mello Jr. view all More in Emerging Tech

Last week, I was at Intels Vision conference at the Virgin Hotel in Las Vegas. While the venue was a bit problematic, the content was solid and showcased an empowered Intel that now needs to address some errors my old friend Pat Gelsinger made as CEO. These include turning the office of the CTO back into a visionary organization, empowering the CMO, adequately funding marketing (while fixing MDF), and ending the cycle of layoffs that has disrupted operational stability.I should point out that I remain impressed with Intels employees, who have weathered the prior storms and still performed in the face of this adversity. They are the unsung heroes of Intels survival.This week, Ill share some thoughts on Intels renewed direction under its new CEO. Then Ill close with my Product of the Week: Linux. The Linux presentation at the end of the second-day keynote reminded me that I was anything but a fan of the platform to begin with. However, it has matured into a true power in the industry, and The Linux Foundation has long been an effort that is both performant and praiseworthy.Intel Vision 2025 Delivers on Its PromiseWhen Intel Vision replaced Intels prior IDF conference, it seemed to lack vision. That changed this year. The 2025 keynote, delivered by Intels new CEO, Lip-Bu Tan, was a landmark event that showcased the companys ambitious roadmap for the future of computing. More than just a presentation of technological advancements, it was a demonstration of a company reinvigorated, led by a CEO uniquely positioned to drive Intels resurgence.Lip-Bu Tans ascension to the helm of Intel has been met with considerable optimism, and rightfully so. Unlike many new CEOs who require a significant period to acclimate, Tan arrives with a profound understanding of Intels inner workings. His prior tenure on Intels board gave him invaluable insights into the companys strengths, weaknesses, and opportunities. Furthermore, his proven track record as a successful tech industry CEO lends credence to his ability to navigate the complexities of the semiconductor landscape. This combination of institutional knowledge and executive experience makes him an exceptionally well-prepared leader for Intel at this critical juncture.The Vision 2025 keynote served as a powerful testament to this preparedness. Tan articulated a clear and compelling vision for Intels future, characterized by a relentless focus on innovation, strategic partnerships, and a commitment to regaining its edge in innovation.Intels Holistic Approach to AIA central theme of the keynote was Intels comprehensive approach to artificial intelligence. Tan emphasized that AI is not merely a feature but a fundamental transformation of computing, impacting every aspect of technology. Intels strategy, as he outlined, centers on providing a full stack of solutions, from silicon to software, to empower developers and enterprises to build and deploy AI applications at scale.This AI Everywhere strategy encompasses:AI-optimized hardware: Intel is developing a diverse portfolio of processors, accelerators, and interconnect technologies specifically designed for AI workloads. This includes advancements in CPUs, GPUs, and specialized AI chips, ensuring customers have the right tools for any AI task.Open software platforms: Intel is committed to fostering an open ecosystem for AI development. This involves contributing to open-source projects and providing robust software tools and libraries that simplify AI development and deployment.Ecosystem collaboration: Recognizing the complexity of the AI landscape, Intel is forging strategic partnerships with leading software providers, cloud service providers, and system integrators to deliver complete AI solutions.Lip-Bu Tan highlighted specific advancements in Intels product roadmap, including:Gaudi 3: The next generation of Intels Gaudi AI accelerators promising significant performance and efficiency gains for deep learning workloads.Falcon Shores: A new high-performance computing (HPC) accelerator designed to bridge the gap between AI and HPC, enabling the convergence of these two critical domains.Xeon processors with AI acceleration: Integrating AI acceleration capabilities directly into Intels flagship Xeon processors, making AI accessible to a broader range of applications.Embracing the Chiplet RevolutionAnother key pillar of Intels Vision 2025 is its embrace of the chiplet architecture. Tan underscored that the traditional monolithic approach to chip design is reaching its limits and that chiplets smaller, specialized chip components are the future of semiconductor innovation.Intels pioneering work in chiplet technology, including its Universal Chiplet Interconnect Express (UCIe) standard, enables the company to deliver more complex and versatile products. By disaggregating chip functions into smaller, reusable chiplets, Intel can:Accelerate innovation: Chiplets allow for faster development cycles, as individual components can be designed and tested independently.Increase flexibility: Chiplets enable the creation of highly customized solutions tailored to specific application requirements.Reduce costs: Chiplets can improve manufacturing yields and reduce development costs by reusing proven components.Tan emphasized that Intel is not only adopting chiplet technology internally but also actively promoting its adoption across the industry through the UCIe consortium. This open ecosystem approach will foster greater interoperability and accelerate the pace of innovation in the semiconductor industry.Intel Doubles Down on Chip ManufacturingLip-Bu Tan reiterated Intels unwavering commitment to manufacturing leadership. He stressed that owning and controlling its manufacturing capabilities is crucial for Intel to maintain its competitive edge and ensure a secure customer supply chain. Intels IDM 2.0 strategy, which combines internal manufacturing, external foundry partnerships, and Intel Foundry Services (IFS), is designed to achieve this goal. Tan highlighted the significant investments Intel is making in new fabs and process technologies, positioning the company to meet the growing demand for semiconductors.Intel Foundry Services (IFS) is a critical component of Intels strategy, aiming to become a leading provider of foundry services to other companies. Tan articulated a vision for IFS to become the foundry of choice for the AI era, offering specialized capabilities and technologies to support the development of next-generation AI chips.Strategic Partnerships: A Force MultiplierTan underscored the importance of strategic partnerships in achieving Intels vision. He emphasized that the computing landscapes complexity requires collaboration across the industry to deliver complete solutions.Intel is actively collaborating with a wide range of partners, including:Software providers: Working with companies like Microsoft, Google, and VMware to optimize software for Intel hardware.Cloud service providers: Partnering with AWS, Azure, and Google Cloud to enable seamless deployment of Intel-based solutions in the cloud.System integrators: Collaborating with companies like Dell, HP, and Lenovo to deliver complete systems and solutions to end users.These partnerships are essential for Intel to extend its reach and impact, ensuring that its technologies are accessible and easy to use for customers across various industries.A Resurgent Intel Under New CEOThe Intel Vision 2025 keynote was more than just a presentation of technology; it was a declaration of Intels resurgence. Under the capable leadership of Lip-Bu Tan, Intel is embarking on a bold new vision characterized by a relentless focus on innovation, strategic partnerships, and a commitment to reclaiming its technological leadership.Tans unique combination of experience and vision makes him an exceptionally well-equipped leader to guide Intel through this transformation. His deep understanding of the semiconductor industry and his proven track record of success position him to navigate the challenges and capitalize on the opportunities that lie ahead.The Vision 2025 keynote provided a compelling glimpse into Intels future, and the industry is watching with anticipation as Lip-Bu Tan and his team execute this ambitious roadmap. Intel is not just talking about the future of computing; it is actively shaping it.Intel Vision 2025: Key Takeaways and AnalysisSeveral key takeaways from the keynote warrant further analysis:AI as a pervasive force: Intels AI Everywhere strategy recognizes that AI is not a niche technology but a fundamental driver of innovation across all sectors. This holistic approach positions Intel to capitalize on the explosive growth of the AI market.Chiplets as a game-changer: Intels leadership in chiplet technology is poised to revolutionize chip design, enabling greater flexibility, customization, and performance a crucial differentiator for Intel in an increasingly competitive market.Manufacturing as a strategic advantage: Intels commitment to manufacturing leadership is essential for ensuring a secure supply chain and maintaining control over its technology roadmap. IFS has the potential to become a significant revenue driver for Intel.Partnerships as a necessity: The emphasis on strategic alliances reflects the recognition that no single company can address the complexities of the modern computing landscape alone. Collaboration is essential for driving innovation and delivering complete solutions.CEO leadership: Lip-Bu Tans preparedness and strategic vision are key to Intels success. His ability to articulate a clear roadmap and inspire confidence in Intels future is crucial for driving the companys transformation.Wrapping Up: Intels Next ChapterThe Intel Vision 2025 keynote was a powerful demonstration of Intels renewed focus and ambition. With Lip-Bu Tan at the helm, Intel is embarking on a bold journey to redefine the future of computing.The companys commitment to innovation, strategic partnerships, and manufacturing excellence positions it to reclaim its leadership role in the technology industry. The industry is keen to see how Intel will execute this ambitious vision.LinuxLinux and I had a very rocky start. I felt it was covering up problems rather than addressing them in the early 2000s. When I pointed that out, my life was threatened, one of my pets was stolen (and allegedly killed), and I resigned from Forrester, which turned out to be a good thing.I did go to war for a bit, but I also respected that the legitimate leaders of Linux calmed everyone down. I lived, and I was stronger for the experience and I really didnt enjoy working for Forrester in the first place.Since then, Ive come to respect Linux as a true power and companies like Red Hat as legitimate providers of powerful open-source solutions that were surprisingly high-performing and robust. Even Microsoft, over this period, came to respect and embrace Linux; more recently, it has become the unsung hero of the AI revolution.In the whirlwind of AI advancements, one foundational element often remains in the background, yet its impact is undeniable: Linux. Under the steadfast leadership of Jim Zemlin at The Linux Foundation who spoke at the Intel event and triggered this Product of the Week showcase this open-source operating system has quietly become the bedrock upon which the most cutting-edge AI technologies are built.While the recent market tremors caused by DeepSeeks emergence momentarily shook investor confidence, a more profound narrative is unfolding. This is a story of open-source collaboration, strategic foresight, and the democratization of AI. Linus Torvalds, the visionary behind Linux, continues to contribute his brilliance, albeit with the autonomy that fosters true innovation.Linux Powers the Future of AIThe shift toward inference, as AI transitions from research labs to mainstream applications, underscores the practical, real-world relevance of Linux. Its flexibility and robustness make it an ideal platform for deploying AI models at scale. Sam Altmans announcement of a forthcoming open-source model from OpenAI further solidifies this trend, highlighting the growing recognition that open collaboration fuels rapid progress. DeepSeeks impressive capabilities are a testament to the long-term strategic vision of embracing open-source technology. While some may perceive Chinas rapid advancements as a competitive threat, its essential to recognize that its approach is largely rooted in internal competition and a deep understanding of open-source principles, many of which were learned from the United States. The intense competition within Chinas tech sector, coupled with a focus on open source, has propelled them forward.Its a misconception to view open source as a barrier to competitive advantage. In todays interconnected world, its a powerful vehicle for achieving it. Intels decisive victory over Sun Microsystems proprietary platforms was largely attributed to its embrace of open-source strategies. Even Apple, a company synonymous with proprietary ecosystems, relies heavily on open-source software.The Engine of InnovationThe reality is that most of the code underpinning modern products and services is derived from open-source projects. With its vast community and continuous development, Linux stands at the forefront of this movement. Its the invisible force driving innovation, enabling developers to build upon a solid foundation and accelerating the pace of technological advancement.Under Jim Zemlins guidance, The Linux Foundation has cultivated an ecosystem that fosters collaboration, transparency, and shared progress. Its a testament to the power of collective intelligence, where contributions from individuals and organizations around the globe coalesce into a robust and ever-evolving platform.As AI continues to shape our world, Linux will remain an indispensable component, quietly powering the next generation of breakthroughs. Its open-source nature ensures accessibility, adaptability, and continuous improvement, making it the ideal foundation for a future where AI is seamlessly integrated into our lives.The Linux story is one of shared success, a testament to the power of open collaboration, and a reminder that the most profound innovations often arise from collective effort, so its my Product of the Week.

Sourcetable AI Spreadsheet Aims To Democratize Data AnalyticsBy John P. Mello Jr.April 1, 2025 5:00 AM PTImage Credit: Sourcetable ADVERTISEMENTAchieve Financial Clarity with SettleGetting accurate product cost data is crucial for growth. Settle unifies invoices, payments, and inventory to help e-commerce operators uncover true costs and boost margins -- all in one platform. Get Started Today! During the dawn of the PC era, a single program was touted as the killer app that fired the market for personal computers. That program was a spreadsheet called VisiCalc. Now, another spreadsheet could have a similar impact on the AI market.Announced Monday, along with US$4.3 million in venture funding, the new spreadsheet, cloud-based Sourcetable, promises to use AI to democratize data analysis for everyone. While 750 million people use spreadsheets every day, only 20% know how to use basic analytical functions, according to Sourcetable.It explained in a statement that the AI spreadsheet eliminates the technical barrier that has plagued spreadsheets since their inception. Users simply tell the spreadsheet what they want done through natural language commands, and Sourcetables AI does the work for them. Instructions can be made via keyboard or through a hands-free voice control mode.Sourcetable added that the spreadsheets autopilot mode can complete a wide range of complex tasks that typically require advanced spreadsheet knowledge, including creating and editing financial models, generating spreadsheet templates, building pivot tables, cleaning data, creating charts and graphs, editing formatting, enriching data, and analyzing entire workbooks.In addition, the AI can understand data context without requiring users to pre-select ranges, interpret multiple ranges across different tabs, work with messy data, and seek human clarification when instructions are unclear.Smart AI Analyst On DemandWe make it easier for executives and managers to have a lot more digital intelligence at their disposal to do tasks, said Sourcetable CEO Eoin McMillan. We also make it easier for junior employees or recent grads to upskill very, very quickly.And we make it easier for people, whether theyre in marketing or sales or accounting or finance, to do both complex workflows, which may not have been possible before based on their skills and training, as well as save them a lot of time on simple workflows, which can be manually intensive, he told TechNewsWorld.Eoin McMillan, CEO, and Andrew Grosser, CTO, are the co-founders of Sourcetable. Photo by Sophia Morel.McMillan explained that partnering with AI is like having a brilliant digital analyst on demand. It allows you to do harder things easier and time-consuming tasks faster. Workflows that took hours can take minutes.Chris Sorenson, CEO of PhoneBurner, a power dialer SaaS company in Seattle, maintained that bringing AI into spreadsheets is a logical evolution. Spreadsheets are powerful, but traditional spreadsheets require a high level of expertise to manipulate large datasets, build formulas, and extract insights, he told TechNewsWorld. As someone with a sales background, I speak to this from experience.With AI, we are seeing a shift people can ask natural-language questions and get real answers without needing to be a data analyst, he said. Thats a major productivity unlock, especially for teams that dont have dedicated data pros on hand.I do think Sourcetables idea that everyone can become an analyst has merit, he continued. It lowers the barrier to entry for data-driven decision-making. But theres still a learning curve in understanding what the AI is doing behind the scenes and theres always a risk of misinterpreting data if you take results at face value without understanding the context. I really believe this could be similar to what VisiCalc did for PCs, he asserted. If AI can better turn spreadsheets into something smarter and more accessible, it could become a gateway to broader AI adoption across departments, but time will tell.Mark N. Vena, president and principal analyst with SmartTech Research in Las Vegas, agreed. Just as VisiCalc and Lotus 1-2-3 brought computing to a broader audience by solving a practical need for accounting and planning, AI spreadsheets could do the same for AI, he told TechNewsWorld. They package powerful capabilities into a familiar interface, making AI accessible to a much wider group of users.Killer App or Spreadsheet Killer?Robin Patra, head of data, analytics, and AI at a large U.S. contractor that relied heavily on Excel workflows before moving to Sourcetable, maintained that just as VisiCalc and Lotus 1-2-3 popularized PCs, AI spreadsheets could catalyze mainstream AI adoption.VisiCalc transformed spreadsheets from paper-based ledgers to digital tools, driving Apple II sales, he told TechNewsWorld. Similarly, Sourcetables self-driving interface represents a paradigm shift.VisiCalc also enabled non-programmers to compute finances, he added. Sourcetable empowers non-technical users to execute advanced analytics. While Sourcetable could be a killer AI app, it could also be a spreadsheet killer. AI will likely make spreadsheets as we know them obsolete because it can take direction and potentially create better outcomes without needing the spreadsheet technology to arrive at this better outcome, explained Rob Enderle, president and principal analyst at the Enderle Group, an advisory services firm in Bend, Oregon.You just tell the AI what you need, input the data, and then get the improved results, he told TechNewsWorld.He cautioned that AI can have accuracy problems even with math. AI can introduce hard-to-identify errors, making the results unreliable, he warned.How AI Spreadsheets May Shape the FutureTo be clear, Vena added, one non-trivial downside is over-reliance on AI-generated output, which may not always be accurate or explainable.Theres also the risk of reduced skill development in traditional analysis methods and potential privacy concerns when data is processed through AI systems, he said.Sourcetables contention that AI can make everyone an analyst needs to be examined critically, he added. AI can level the playing field by enabling non-experts to run analyses and extract insights, he noted, but Sourcetable assumes that AI always delivers useful or correct answers which isnt guaranteed. Users still need judgment to interpret results, so like any AI technology, its not fail-safe.Looking forward, Sourcetable could influence not only spreadsheet applications but also consumer applications in general. AI spreadsheets could phase out roles centered on manual data entry, shifting demand toward AI management skills, Patra said.Natural language interfaces could also start to replace traditional GUI-based tools, he added.AI-powered consumer tools like Sourcetable could reshape productivity software, Vena said. They may lead to faster decision-making, more empowered individuals, and broader adoption of data-driven thinking.However, he continued, they also raise issues around data security, job displacement, and trust in automated systems. But putting those issues aside, this has the potential as a real breakthrough that both businesses and mainstream consumers could benefit from.John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John.Leave a CommentClick here to cancel reply. Please sign in to post or reply to a comment. New users create a free account.More by John P. Mello Jr.view allMore in Artificial Intelligence

Nvidias GTC (GPU Technology Conference) has firmly established itself as the leading event for artificial intelligence (AI) and accelerated computing. GTC 2025 was no exception. It was packed with significant announcements and showcased the companys continued dominance in the AI hardware space. However, the event also highlighted some potential challenges and shifts in the competitive landscape.Lets break down the key announcements from GTC 2025. Then, Ill close with my Product of the Week a standout AI laptop from Lenovo that reflects where PC innovation is heading.GTC 2025: Key AI and Hardware RevealsGTC 2025 saw Nvidia unveil a range of new products and technologies, reinforcing its position at the forefront of AI innovation. Key announcements included:Blackwell Ultra: Nvidia introduced the Blackwell Ultra GPU architecture, an enhanced version of its Blackwell platform designed for demanding AI reasoning models. Blackwell Ultra boasts increased memory capacity and improved performance, further solidifying Nvidias commitment to pushing the boundaries of AI compute.Rubin Architecture: Looking ahead, Nvidia announced the Rubin architecture, the successor to Blackwell, promising even better performance and efficiency for AI data centers. This announcement underscores Nvidias annual cadence of innovation in AI infrastructure.Isaac GR00T N1: Nvidia made a significant push in robotics with the unveiling of Isaac GR00T N1 (pictured above), the worlds first open, fully customizable foundation model for humanoid robots. This initiative aims to accelerate the development of general-purpose robots capable of performing a wide range of tasks.Quantum Computing Efforts: Nvidia announced its foray into quantum computing by establishing the Nvidia Accelerated Quantum Computing Research Center (NVAQC) in Boston. This initiative signals Nvidias intention to play a key role in the future of quantum computing.These announcements demonstrate Nvidias continued focus on driving advancements across the AI spectrum, from fundamental hardware to software platforms and emerging fields like robotics and quantum computing.Risks of Nvidias AI Market LeadershipNvidias dominant position in the AI hardware market also carries inherent risks. The danger of being the leader in any field is the potential to become complacent and take competition for granted. While Nvidia continues to innovate, there were aspects of GTC 2025 that raised some concerns. Notably, the event lacked the compelling demonstrations of real-world AI applications that characterized previous GTCs. In past years, GTC showcased how Nvidias technology was being used to solve specific problems and drive innovation across various industries.This year, with the exception of robotics, the focus seemed to be more on the underlying technology and future roadmaps than on tangible, present-day applications.Missed Opportunity for AI in ActionWhile the robotics demonstrations were impressive, they often leaned toward entertainment rather than showcasing practical applications for getting real-world work done. For example, while seeing a robot out of Star Wars is fun to watch, it may not have resonated as strongly with professionals seeking to leverage AI for business productivity or scientific research.Furthermore, Nvidias CEO, Jensen Huang, a charismatic and influential figure in the tech industry, did not prominently feature a personal AI assistant or demonstrate how he personally uses AI to enhance his daily workflow.In an era where AI assistants should be becoming increasingly sophisticated, the absence of such a demonstration from the leader of an AI powerhouse was notable. This absence raises questions about the maturity and practicality of AI assistants in everyday executive use, or at least Nvidias public messaging around them.Growing Threats to Nvidias AI DominanceNvidias dominance in the AI hardware market is also facing increasing competition:AMD has emerged as a strong number two in the GPU market, making significant inroads into Nvidias traditional strongholds. AMD offers competitive products and solutions, particularly in the data center space, and is gaining traction with key customers.China is rapidly developing its own AI hardware capabilities. Companies like Huawei and others are investing heavily in domestic GPU and AI chip development, driven by both commercial ambitions and a desire for technological self-reliance. Given the ongoing geopolitical tensions and trade restrictions, Chinese companies are highly motivated to create competitive alternatives to Western technology, including Nvidias products.A Quantum Leap?Nvidias increased focus on quantum computing, highlighted by the announcement of its new research center, represents a potentially transformative shift for the company. While still in its early stages, quantum computing holds the promise of revolutionizing computation and solving problems that are intractable for classical computers.However, Nvidia is entering a field with already established players. Companies like IBM, Google, and numerous startups are actively pursuing quantum computing research and development. China is also making significant investments in this area, viewing it as a strategic technology for future economic and technological leadership. Other companies with compelling quantum efforts include:Rigetti ComputingHoneywell Quantum SolutionsIonQPsiQuantumNvidias expertise in accelerated computing could give it an advantage in developing hybrid quantum-classical systems, but it faces stiff competition from these established players.Gaming Takes Back Seat at GTCAnother notable aspect of GTC 2025 was the relative lack of emphasis on gaming.In the past, GTC was a significant event for showcasing the latest advancements in gaming technology, with a strong focus on graphics performance and new gaming experiences. However, gaming took a backseat to AI and data center technologies this year. The absence of significant announcements related to AI-powered non-player characters (NPCs) in games was conspicuous. While AI has the potential to revolutionize gaming by creating more dynamic, realistic, and engaging virtual characters, this area did not seem to be a major focus at GTC 2025.Wrapping UpNvidia is undoubtedly at a pivotal moment. The companys leadership in AI hardware has propelled it to unprecedented success, but it cannot afford to rest on its laurels. To maintain its position and capitalize on the full potential of AI, Nvidia must:Demonstrate Real-World Impact: Focus on showcasing how its technology is being used to solve practical problems and drive tangible benefits across industries.Embrace the Application Layer: Go beyond simply providing the hardware and play a more active role in enabling the development of innovative AI applications.Navigate the Competitive Landscape: Aggressively address the challenges posed by rising competitors like AMD and Chinas growing technological prowess.Strategically Pursue New Frontiers: While exploring promising areas like quantum computing is essential, Nvidia must ensure that these efforts complement, rather than distract from, its core focus on AI.Remember Its Roots: While the company has expanded far beyond its gaming origins, Nvidia should not completely abandon this market, as it remains an important source of innovation and a key driver of technological advancement.GTC 2025 underscored Nvidias continued leadership in AI hardware but also highlighted the challenges and opportunities that lie ahead. To thrive in the long term, Nvidia must remain focused, demonstrate the practical power of AI, and avoid devolving into a company that merely touts speeds and feeds.Lenovo ThinkPad X9 14 Aura EditionPhoto by AuthorThe Lenovo ThinkPad X9 14 Aura Edition represents a change of direction from traditional ThinkPad design, offering a modern, ultraportable take on the business laptop. It also showcases how far Intel has come in the AI PC race. The X9 14 Aura Edition sports a sleek, slim, and light design, a departure from the classic matte black hybrid chassis of previous ThinkPads.Design and Build QualityIts constructed from recycled aluminum and comes in a Thunder Gray color. It has premium build quality and feels modern, with tiny display bezels and a sizeable haptic touchpad. To achieve its slim profile, Lenovo incorporated an engine hub bar at the bottom rear of the laptop, housing expansion ports and the cooling system. This design is functional and gives the X9 a unique look.The laptop is also MIL-STD 810H certified, indicating it can withstand travel hazards like shock, vibration, and extreme temperatures. Its unique ribbed bottom makes it one of the easiest Ive tested to hold and carry.Display Options and Visual ExperienceThe X9 14 Aura Edition features a 14-inch display. A key highlight is the option for a 2.8K OLED touchscreen, which is described as bright and colorful and offers excellent contrast. The OLED display supports Dolby Vision and Windows 11s adaptive color capabilities. Other configurations include a 1920 x 1200 resolution non-touch display.The laptop is designed for efficiency and extended power. In terms of performance, the device is powered by Intel Core Ultra processors and includes integrated Intel Arc graphics. Configurations come with LPDDR5X memory. Storage is provided by M.2 PCIe Gen4 SSDs.Performance and AI CapabilitiesThe X9 14 Aura Edition includes AI features co-developed with Intel, such as Smart Modes like Attention and Wellness. It has a high-resolution 8 MP webcam, which is noted for its quality. It includes Thunderbolt 4 ports, an HDMI port, and Wi-Fi 7 for connectivity. Security features include a fingerprint reader in the power button and Windows Hello face recognition.Keyboard, Touchpad, and AudioThe keyboard is an island-style keyboard with soft, flat, square keys. Some reviews note that it has a shallower feel compared to the keyboards on other ThinkPad models. The laptop has a large, buttonless touchpad that glides smoothly.In terms of audio, the X9 14 Aura Edition includes downward-firing stereo speakers with Dolby Atmos spatial audio capabilities. The audio quality is described as clear and sufficient to fill a room.Bottom Line on the ThinkPad X9 14 Aura EditionThe X9 14 Aura Editions pros include its compact and modern design, premium build quality, gorgeous OLED display option, and modern connectivity features. Its cons include a keyboard that may not be Lenovos best, a lack of USB-A ports and an SD/microSD card slot, and no mobile broadband option in some configurations.Overall, the Lenovo ThinkPad X9 14 Aura Edition is a modern and stylish ultraportable laptop that blends performance, portability, and a premium user experience. This business laptop is a departure from traditional ThinkPad designs, targeting users who prioritize a sleek and modern aesthetic without sacrificing business-grade features and it earns my pick for Product of the Week.

Apple, why have you forsaken users like me? I feel like youre beginning to take me for granted.I know that Apple is still one of the most influential tech companies on the planet, and I fully recognize that its products continue to set the bar in industrial design, brand loyalty, and ecosystem integration.But not everything is perfect in Cupertino.Despite Apples dominance, there are discernible cracks in the experience: frustrations that long-time users and analysts cant ignore. Not to mention that Samsung recently extracted me from the Apple ecosystem with its new Galaxy S25 Ultra.While Apple continues to lead in many areas, aspects of its ecosystem have begun to show signs of wear. Here are four specific gripes that detail missed opportunities and areas for improvement for Apple in 2025.1. The Apple Watch Feels StagnantWhen Apple Watch launched in 2015, it was a bold leap into wearables. But nearly a decade later, the product feels stuck in place.Battery life remains a glaring issue. Most baseline models still need nightly charging, which is a massive problem for a device that needs to be worn on the body and is marketed as a health monitor and sleep tracker.Even in the 2024 Series 9 and SE models, Apple hasnt pushed the battery life boundary forward. Multi-day wear standard in devices from competitors like Garmin or even budget Fitbit models remains out of reach. The design language hasnt meaningfully evolved either. While Ultra models added bulk and utility, the mainstream Apple Watch still looks and feels like a product from five years ago.The market reflects this fatigue. According to a recent report from Counterpoint Research, global smartwatch shipments fell in 2024, with Apple losing share. The data suggests user enthusiasm is cooling and Apple hasnt responded with the bold rethinking it once championed.Heres a provocative question: What if Apple treated the Watch as a health-first platform instead of just a phone accessory?2. Apple Intelligence Isnt Delivering a BreakthroughThere were high expectations when Apple announced its push into generative AI, dubbed Apple Intelligence. This point isnt only my opinion, as Apple has been under fire from the media for its sluggish ability to make Apple Intelligence resonate with mainstream consumers. Not to mention that Apple changed the management leadership overseeing this capability within the company.Apple has not identified a standout usage model that resonates with a broad consumer base. Theres no ChatGPT moment. Instead, Apples AI ambitions feel vague and scattered. Features like auto-summarizing messages, writing assistance, or Siri enhancements are interesting but incremental.Saying Apple Intelligence hasnt lit the world on fire may be an understatement.Unlike Microsoft, which integrates AI deeply into the productivity stack, or Google, which is aggressively building Gemini into its services, Apple hasnt identified a killer use case. Nor has it clearly positioned Apple Intelligence as a system-wide differentiator.Thats problematic.Consumers are increasingly aware of AI but need to see tangible value. Apples AI story currently lacks urgency, clarity, and consumer pull. The company needs the proverbial killer app (or at least usage model) that gets mainstream consumers on board. Remember Lotus 1-2-3?The bottom line is that Apple needs to make AI feel indispensable to everyday users not just a list of background features.3. The iPhoneWindows Experience Remains PainfulIn 2025, cross-platform integration should be seamless.Yet, if you use an iPhone and a Windows 11 PC, it still feels like two worlds that refuse to talk. This issue is, perhaps, my biggest peeve with Apple.Basic tasks like message syncing, photo transfers, and notifications often involve clunky workarounds. Theres no true equivalent to iMessage or AirDrop on Windows even though nearly half of U.S. users run Windows PCs.Yes, Apple and Microsoft have made minor progress. Features like iCloud integration in the Photos app or the recent Phone Link improvements are a step forward. But theyre patchwork solutions, not a cohesive strategy. In contrast, Android users get a far smoother experience with Windows out of the box thanks to better file sharing, notifications, and cross-app integration.Apple may want to keep its ecosystem closed. However, doing so risks alienating a massive segment of its user base that lives in both the Apple and Microsoft worlds.The question worth asking is whether the walled garden is still a strength or is becoming a liability in a multi-device/multi-brand era.4. Apple Is Behind on Foldable PhonesFoldables are no longer a niche experiment. Samsung has iterated rapidly. Google, OnePlus, and Huawei are pushing new designs. Even midrange foldables are entering the mainstream.Yet, Apple still hasnt entered the race.Despite years of rumors, weve seen no tangible movement on a foldable iPhone. There are no prototype previews, no developer hints, and no firm timeline.Thats puzzling. While Apple isnt usually the first to enter a category, it usually sets the standard when it does. The lack of any real foldable roadmap raises questions about Apples willingness to take hardware risks.Some reports suggest Apple is experimenting behind the scenes. But thats not enough. The market is maturing now, and users expect new form factors that offer flexibility, multitasking, and larger screens in compact devices.By waiting too long, Apple risks looking like a follower rather that an innovator.The hard truth is that Apple risks missing a generational hardware shift if it doesnt show leadership here.WWDC 2025: A Chance To Reset the NarrativeApple still has time to correct its course, and its Worldwide Developers Conference in June is the perfect opportunity.These issues wearable fatigue, AI uncertainty, cross-platform friction, and form factor hesitation could be addressed through bold announcements and more precise product direction.Imagine if WWDC delivers:A redesigned Apple Watch with multi-day battery life and new health sensorsA more compelling and intuitive Apple Intelligence platform that redefines mobile AIDeeper iPhoneWindows integration that finally removes the frictionA foldable iPhone concept that sets a new design benchmarkApple is unquestionably capable of all of this. The question is whether it will rethink its product roadmap and move faster.Users are no longer satisfied with just minor updates. Theyre hungry for fresh ideas, more substantial experiences, and openness.WWDC could either mark a turning point or reinforce the growing perception that Apple is slowly losing its edge. So, the big question is: Will Apple rise to the occasion in June?

AI red teaming the practice of simulating attacks to uncover vulnerabilities in AI systems is emerging as a vital security strategy.Traditional red teaming focuses on simulating adversarial tactics to identify vulnerabilities in static systems applications built on predictable coding frameworks. While challenging, these engagements rely on decades of collective experience and a well-understood threat landscape. Artificial intelligence systems, meanwhile, present fundamentally different challenges.Unlike traditional applications, AI systems are dynamic, adaptive, and often opaque, making vulnerabilities harder to identify. The threats targeting these systems are also unique, ranging from adversarial inputs that manipulate model behavior to malicious modifications of the models themselves. Understanding these differences is critical for securing AI-powered operations.Understanding AI Red TeamingAI red teaming systematically probes AI systems to identify vulnerabilities and potential exploits. Like traditional red teams, AI red teams emulate adversarial tactics to uncover weaknesses before they can be exploited. However, the challenges in AI environments are amplified by model complexity, novel attack vectors, and the high stakes of failure.Red teaming spans critical areas. Adversarial machine learning (ML) techniques involve crafting inputs to deceive models into making incorrect predictions. Model file security focuses on identifying vulnerabilities in the serialized files that store machine learning models, ensuring they cant be exploited as vectors for malicious code. Operational security examines risks within AI workflows and supply chains, pinpointing exposure points adversaries might exploit.Why AI Red Teaming Is CriticalThe adoption of AI has transformed industries, introducing significant risks alongside benefits. Models trained on sensitive data, deployed in distributed environments, or integrated into decision-making processes create a broader attack surface. Adversaries can manipulate inputs through vulnerabilities in large language models (LLMs), leading to unintended outputs or actions.AI models, often valuable intellectual property, are also vulnerable to theft or sabotage, leading to severe business impacts. Maliciously crafted models can even act as malware, leaking sensitive data when deployed.For example, an attacker might embed harmful code within a serialized model file. If a machine learning engineer unknowingly loads the file, the attacker could gain access to sensitive systems. This threat is particularly severe in industries like healthcare and finance, where compromised models could directly impact operations and compliance.Modernizing Red Teaming for AITraditional cybersecurity measures often fail to address AI-specific threats. Unlike fixed codebases, AI models are dynamic and adaptive, introducing unpredictable vulnerabilities that require innovative testing strategies.AI attack vectors often involve reverse engineering or adversarial attacks that undermine predictions. Additionally, tools designed to secure AI systems can sometimes be weaponized for offensive use, requiring a deep understanding of their dual-use capabilities. These challenges underscore the necessity of AI red teaming as a specialized cybersecurity discipline. By focusing on AI systems unique vulnerabilities, red teams can identify risks that traditional approaches may overlook, ensuring a more robust defense.Given the growing use of AI in business-critical applications, AI red teaming is emerging as an essential cybersecurity strategy. This process requires continuous adaptation to evolving threats. Integrating AI red teaming into an organizations security framework reduces catastrophic model failures, builds stakeholder confidence in AI initiatives, and keeps organizations ahead of adversaries.Best Practices To Get Started With AI Red TeamingTo integrate AI red teaming into an enterprise security strategy, consider these best practices for building a robust and proactive approach to safeguarding AI systems.Assemble a Multidisciplinary TeamAn effective AI red team requires AI experts to address model architecture and vulnerabilities, cybersecurity professionals to tackle adversarial tactics, and data scientists to analyze risks like poisoning or unauthorized manipulation. This combination ensures a comprehensive approach to securing the AI lifecycle.Identify and Prioritize Critical AI Assets for TestingFocus on assets that handle sensitive data, perform critical functions, or are publicly accessible. Catalog models, data pipelines, and APIs, prioritizing systems like fraud detection or customer authentication where vulnerabilities could have significant consequences.Collaborate With Blue TeamsRed and blue teams must work together to strengthen defenses. While red teams simulate attacks to uncover vulnerabilities, blue teams focus on detection, response, and mitigation. Sharing findings from red teaming helps improve blue team capabilities and align strategies against real-world threats.Use Specialized ToolingEquip your red team with tools tailored to AI environments, including frameworks for adversarial inputs, reverse engineering model embeddings, or simulating API attacks. These tools help target AI systems more effectively and uncover weaknesses that traditional security testing might miss.Implement Automated TestingFor organizations managing large-scale AI deployments, automated red teaming tools streamline vulnerability detection across multiple models and pipelines, enabling comprehensive coverage without overburdening teams.Ensure Compliance With Privacy GuidelinesEstablish protocols to ensure red teaming activities meet privacy and regulatory standards. This includes anonymizing sensitive data in testing environments to safeguard personal and proprietary information. Proper compliance not only protects your organization legally but also ensures ethical testing practices.Making Red Teaming Part of AI StrategyBy adopting a proactive approach to AI security with red teaming, businesses can uncover hidden vulnerabilities, reduce risks, and build resilient systems. Compared to static applications, the dynamic nature of AI systems makes offensive testing even more important for safeguarding operations, maintaining stakeholder trust, and driving innovation with confidence.

Screen Time of Americans Above Global Average: StudyBy John P. Mello Jr.March 26, 2025 5:00 AM PT ADVERTISEMENTB2B Online Chicago Get in the GameGreat teams win with great plays. At B2B Online Chicago, youll get the strategies top eCommerce and digital marketing pros are using to drive growth. Dont sit on the sidelines - get in the game. Use code B2BCHI25PARTNER for 20% off. Americans spend an average of six hours and 40 minutes a day staring at the screens of their computers, phones, and other electronic devices.If that sounds like a lot, it is, but its only two minutes above the global average, according to a worldwide screen time study released Wednesday by Comparitech, a reviews, advice, and information website for consumer security products.While staring at screens for more than six hours a day may sound excessive, experts appear conflicted on the subject. It is hard to define what is excessive when it comes to screen time, especially as many of us rely on the use of screens for work, said Comparitech Head of Data Research Rebecca Moody.However, where things perhaps do become excessive is in their mobile screen time which isnt likely to come from work-related activities for the majority of people, she told TechNewsWorld.According to the study, Americans spend three hours and 22 minutes daily staring at mobile screens, slightly below the global average of three hours and 46 minutes.Titania Jordan, chief parent officer at Bark a parental control and monitoring solutions company in Atlanta and author of Parental Control: A Guide to Raising Balanced Kids in the Digital Era, explained that every person is different and depending on their work type, the amount of screen time will vary dramatically.One thing to keep in mind when gauging whether youre spending too much time on screens is if its interfering with other aspects of your life and how old you are, she told TechNewsWorld.Are you foregoing real-life hangouts? she asked. Is your work or school suffering? Are you getting enough quality sleep? Do you feel more anxious or depressed after doom-scrolling? All these factors come into play when figuring out your personal screen-time sweet spot.Is 6 Hours of Screen Time Too Much?To determine if an amount is too much or too little, we need to know what the damages and benefits to the individual are, added Rob Enderle, president and principal analyst at the Enderle Group, an advisory services firm.If this is making them less healthy, which I expect it likely is by causing eyesight degradation, issues resulting from the lack of exercise, socialization issues, and creating excessive political polarization then, yes, it is way too much time, he told TechNewsWorld.But there could be offsetting benefits, especially when facing multiple potential pandemics, as the isolation keeps them safer, and some may be more productive that way, he continued.In the end, though, for most, I expect this is extremely unhealthy, he added. Six hours of screen time is absolutely excessive for most people, argued Mark N. Vena, president and principal analyst with SmartTech Research in Las Vegas. Six and a half hours is nearly a third of the day, excluding sleep and work, much of it spent passively consuming content, he told TechNewsWorld.While some screen use is productive or relaxing, large daily totals suggest a displacement of healthier habits like exercise, in-person interaction, and sleep, he added. The concern isnt just the time but what that time replaces, which is something parents should focus on with their kids.Larry D. Rosen, professor emeritus at California State University, Dominguez Hills, who has written extensively on the psychology of technology and screen time, agreed that six-plus hours of screen time is excessive. It takes away from time spent interacting with people face to face, as well as limiting time spent doing non-screen activities, such as exercise or just allowing our brain to wander, he told TechNewsWorld.He added that the data hes gathered on mobile phone usage shows far more screen time than the average in the survey, with the majority spent connecting to others through social media and messaging.According to the survey, Americans spend an average of two hours and nine minutes a day on social media, slightly lower than the worldwide average of two hours and 21 minutes.It noted that the social media platforms that Americans are most fond of include YouTube, which is used by 85% of U.S. adults, Facebook (70%), Instagram (50%), Pinterest (36%), TikTok (33%) and LinkedIn (32%).Screen Time Higher in Developing RegionsOverall, according to the study, South Africans were the biggest screen consumers, spending nine hours and 27 minutes a day on screens. The biggest desktop screen consumers were the Russians, spending four hours and 31 minutes a day on screens, while Filipinos were the largest mobile screen consumers, spending five hours and 21 minutes a day on screens.On the whole, the study noted the biggest screen-time consumers were in Africa, Asia, and South America. [Screens have] become the primary source of news in the non-first-world countries since the Arab Spring, Rosen said.In many of these regions, its important to note that smartphones are the primary gateway to the internet, making screen use a central part of daily life, Vena explained. High youth populations and rapidly growing digital economies contribute to greater mobile engagement, he added. In some cases, there may be fewer accessible offline entertainment or public infrastructure options. Social media and streaming platforms also offer affordable entertainment where traditional media is limited or costly.Jordan noted that certain platforms like TikTok, Instagram, and WhatsApp are extremely popular in these areas, not only for social interaction but also for business exchanges and news consumption.Also, she continued, many of these countries skipped broadband and slower-growth internet completely, jumping quickly to mobile-connected devices and Wi-Fi. This, in turn, meant people in these regions had smartphones first as opposed to desktops and spent more time online due to the mobility of their devices.Global Screen Time Trends Shift DownwardComparitech researchers also found that more than half the countries in the study (27 of 49) reduced their screen time from last year. Hong Kong showed the largest reduction, slicing 50 minutes off its average, dropping from seven hours and 55 minutes to seven hours and 26 minutes.With more countries improving their screen time this year, the most obvious reason for a drop in screen time would be a continuation of normalcy, post-pandemic, the researchers noted.Also, they continued, people have become more aware of the negative impacts lengthy screen time has on their mental health.Furthermore, they added, advancements in technology, such as improved digital well-being features on smartphones and applications, may have contributed to the decrease in screen time by encouraging users to monitor and limit their screen usage.Many of us are becoming increasingly aware of the negative impacts of spending too much time looking at a screen and are perhaps consciously unplugging our devices for an hour or so before we go to sleep, Moody said. Alex Ambrose, a policy analyst with the Information Technology and Innovation Foundation, a research and public policy organization in Washington, D.C., noted that nations have tried to regulate screen time.Time-online policies are sometimes used as an attempt to lower screen times, but they are difficult for policymakers to enforce, she told TechNewsWorld. China, for example, instituted a three-hour-per-week limit for children playing video games. While 77% of children reduced their gaming time under this policy, 29% also reported using their parents accounts as a workaround to the time limit.Some platforms have taken voluntary steps to tackle any potential harms, with measures such as allowing users to limit their screen time and see screen-time reports, or notifying them to take a break whenever they have spent a certain amount of time on an app, she continued.Rather than implementing prescriptive policies that hinder user experience and leave little room for innovation, many technology companies instead adopt customizable user and parental control systems such as screen-time controls and prompts to take breaks after using the apps for extended periods of time, she added.John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John.Leave a CommentClick here to cancel reply. Please sign in to post or reply to a comment. New users create a free account.Related StoriesMore by John P. Mello Jr.view allMore in Internet

Crashing the Boys Club: Women Entering Cybersecurity Through Non-Traditional PathsBy John P. Mello Jr.March 25, 2025 5:00 AM PT ADVERTISEMENTB2B Online Chicago Get in the GameGreat teams win with great plays. At B2B Online Chicago, youll get the strategies top eCommerce and digital marketing pros are using to drive growth. Dont sit on the sidelines - get in the game. Use code B2BCHI25PARTNER for 20% off. Although many women take the traditional IT route into cybersecurity, a large portion (37%) have entered cyber from non-IT or military positions, according to a survey released Friday by ISC2, the organization that maintains and administers the Certified Information Systems Security Professional (CISSP) certification exam.The organization reported that a significantly higher percentage of women than men who participated in the same study have leveraged education, professional development, self-initiated experience opportunities, and apprenticeships to pursue their careers.The report also noted that 56% of women respondents said their organizations are already changing their hiring requirements to bring in more people from non-cybersecurity backgrounds. This is illustrative of employer efforts globally to widen the potential cybersecurity talent pool without compromising standards, the report maintained.This is something that we currently observe not only in cybersecurity but in tech roles in general, said ydrn Vitait, business unit manager for Monitum, a software-as-a-service company specializing in smart security solutions with offices in Sweden and Lithuania, and a co-founder of Women Go Tech, an NGO that focuses on empowering women in the tech industry.First of all, women who had previously built careers in different sectors have more diverse backgrounds, stronger problem-solving skills, and advanced pattern recognition skills, she told TechNewsWorld.Another key advantage here is transferable skills, she continued. Such women can work in dynamic environments, where not only knowledge but traits such as curiosity, problem-solving, creativity, and other aspects matter.From our experience in re- and up-skilling programs, we have seen adult women in their thirties to forties deciding to shift from non-tech to tech, and they do it exceptionally successfully, she added.Non-IT Backgrounds Open Cybersecurity DoorsThomas Vick, a technology hiring and consulting expert at Robert Half, a global staffing and recruiting firm, explained that hiring individuals with non-IT backgrounds can often bring diverse perspectives and problem-solving approaches shaped by their unique experiences.For example, there is a growing demand for interpersonal and management skills in IT roles, underscoring the importance of understanding and integrating business methodologies into technological development, he told TechNewsWorld.Professionals with strong communication, critical thinking, and project management experience can often enhance an IT teams effectiveness, he added. Their varied experiences often lead to creative solutions and a more holistic understanding of the work being done.Alyson Laderman, CEO of Akylade, a provider of cybersecurity certifications with offices in Florida, pointed out that it can be difficult for women to gain IT experience because it, too, is a male-dominated field.The problem is worse now because of shrinking opportunities, she added. Theres not as many of those entry-level-type IT positions to get into to be able to start there and then move up to cyber, she told TechNewsWorld. So I think what youre seeing is a shift in the workforce in general, both in terms of gender, but also in terms of opportunity as to IT being a feeder into cybersecurity. Teresa Rothaar, a governance, risk, and compliance analyst with Keeper Security, a password management and online storage company in Chicago, noted that the climate for women in cybersecurity has been showing signs of improvement over the past few years. However, it still faces substantial challenges, she told TechNewsWorld. Significant work still needs to be done to ensure equal opportunities and a supportive environment for women in this industry.Ongoing challenges to womens advancement in cybersecurity include significant underrepresentation, as women remain notably outnumbered in cybersecurity roles despite efforts to close the gap, and men continue to dominate the field, particularly in leadership and technical positions, she said.The workplace culture in many cybersecurity environments can be unwelcoming to women, with issues such as gender bias, lack of recognition, and sometimes a hostile work environment hindering their progress and retention in the field, she continued.Gender pay gaps persist in cybersecurity, with women often earning less than their male counterparts for similar roles, a disparity that is particularly pronounced in tech fields, she added. Women in cybersecurity also face barriers to career advancement, including fewer opportunities for promotions and leadership roles, as well as a lack of support for continuing education and professional development.Education as Pathway Into CyberThe ISC2 report also noted that alongside the strong percentage of candidates leveraging IT experience to get into cybersecurity, advanced education was a significant pathway into these roles, although more so for women than men.Nearly a quarter (24%) of women respondents said they came in with a cybersecurity-related undergraduate degree (compared to 18% of men), and 23% of women (18% of men) entered a cybersecurity role with an undergraduate degree in a field not directly linked to cybersecurity.In addition, 18% of women respondents (12% of men) noted that they held an advanced degree a postgraduate qualification such as a masters or a doctorate in a cybersecurity-related subject before taking up a cybersecurity role. For non-cybersecurity advanced degrees, 16% of women respondents (11% of men) held these qualifications before entering the cybersecurity workforce. Men get hired on potential. Women get hired on proven ability, Laderman argued. Women are almost required to get that higher education, get more experience, to be able to show that they are of the same value of what typically men would be potentially capable of doing.Its clearly not a fair standard, but its something I myself have been through, she added.Vitait noted that women in her organizations programs also tend to have higher education levels and many certificates in various subjects. Unfortunately, this is a consequence of lower self-esteem and willingness to boost confidence and employability with hard knowledge, she said.Women also tend to apply later in the process of a career shift, she continued. They want to build a knowledge base and prove it via formal and informal education. Only later do they apply for the roles usually when they meet 80% to 90% of the requirements. Men tend to apply to the roles much earlier and learn on the go.Value of Cybersecurity CertificationsAnother finding by the ISC2 report was that beyond career and formal education, women place a significant emphasis on certifications and professional development to support their efforts when securing and furthering a cybersecurity career path. Holding a cybersecurity certification before entering their first job in cybersecurity was cited by 18% of women respondents and 16% of men.Because cybersecurity is a relatively new field, certifications are a great way for professionals to transition into it from another field, said Julia Toothacre, a career strategist, coach, and owner of Ride The Tide Collective, a career coaching company based in Claremont, Calif.Mid-career professionals who have experience in another area might benefit from various certifications to increase their chances of getting hired, she told TechNewsWorld. I dont think there is a gendered perspective when it comes to certifications. Its really based on a combination of experience and education, which will vary between everyone male or female. Rob Rashotte, vice president of global training and technical field enablement at Fortinet, a provider of firewalls, intrusion prevention systems, endpoint security, and antivirus programs in Sunnyvale, Calif., added that reexamining and revising education and training requirements for cybersecurity roles is a great place to start to address the existing workforce shortage in the domain.Many organizations are still likely overlooking solid candidates, he told TechNewsWorld. While many companies have diversity hiring goals, we arent seeing hiring numbers increase significantly among women, minorities, and veterans.According to Fortinets 2024 Global Cybersecurity Skills Gap Report, despite 91% of respondents saying they prefer to hire candidates with technical certifications, 71% of organizations require potential new hires to hold a four-year degree. Organizations should be identifying candidates who possess the right soft skills and then using certifications to help them gain cybersecurity-specific knowledge, Rashotte maintained.He added, According to the report, most leaders are open to this approach, with 89% of respondents saying they would pay for an employee to obtain a certification.John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John.Leave a CommentClick here to cancel reply. Please sign in to post or reply to a comment. New users create a free account.Related StoriesMore by John P. Mello Jr.view allMore in Women In Tech

Ive been attending and covering HPs partner events like Amplify for decades, and it has always struck me that HP treats its partners like family.That kind of partner relationship is vastly different from what Ive seen at other companies Ive worked for firms that claim to value partners but often treat them like a burden theyd rather not deal with. HPs approach should be the standard. Instead, it is the exception. Enrique Lores is a unique CEO, and while HP has had some problematic CEOs, Lores embodies the empathy and customer focus of HPs founders.It is rare to see a CEO mingling with attendees, soliciting feedback, and engaging in meaningful conversations at events. Typically, CEOs retreat to the background after a brief appearance on stage, leaving the spotlight to others.However, HPs Amplify partner event was different. It was marked by the unusual presence of HPs CEO, who actively roamed the event floor, interacting with partners and attendees alike. This level of engagement was unprecedented and set the tone for a truly remarkable experience.Lets talk about HP Amplify this week. Then, well close with my Product of the Week, an upcoming notebook from HP that uses the Qualcomm Snapdragon solution correctly which, sadly, is also an exception.HPs Focus on End Users and PartnersThe HP Amplify event was unusual not only because of the CEOs approachability but also because of its extreme focus on end users and partners.Unlike many industry events that primarily showcase engineering feats and product specifications, Amplify prioritized the experiences and needs of its customers and partners.HP demonstrated a commendable dedication to understanding and addressing the real-world applications and challenges faced by those who use its products daily.AI Tools Streamline Customer, Partner ServiceInternally, HP has been pioneering the aggressive use of artificial intelligence (AI).At Amplify, the demonstration of AI tools designed to streamline the process of matching customers and partners with the right products was particularly noteworthy. These AI solutions enhance the efficiency and accuracy of product recommendations, ensuring that the end users receive the most suitable options to meet their needs.This focus on AI extends beyond internal operations, showcasing HPs commitment to leveraging advanced technologies to improve customer satisfaction.PC and Printer SecurityHPs leadership in security was evident across both its PC and printer lines. With a strong emphasis on protecting user data and ensuring safe operations, HP continues to innovate in this critical area.Its unique Quantum Safe printer offerings represent a significant leap forward in safeguarding sensitive information. These printers are designed to address the emerging threats of quantum computing, ensuring they remain secure even as technology evolves.Expanding AI PC Product LinesIn an impressive move, HP announced the introduction of 60 new AI PC products. This significant expansion positions HP at the forefront of the desktop AI PC market.By focusing on desktop AI capabilities, HP is not only catering to current market demands but also paving the way for future dominance.This initiative highlights HPs strategic approach to incorporating AI into its product development, ensuring it remains competitive and innovative.Employee Care: A Legacy of CompassionHPs unique focus on employee care was another highlight of the Amplify event. This mindset is not a recent development but a foundational principle dating back to the companys inception. The companys commitment to its employees is reflected in various initiatives and policies aimed at ensuring their well-being and professional growth. This approach fosters a positive work environment and contributes to the companys overall success.Sustainability and Product ReuseHPs advancements in product reuse and a market-leading focus on sustainability were prominently showcased at Amplify.The company has made significant strides in reducing environmental impact through innovative practices and sustainable product designs. HPs dedication to sustainability is evident in its efforts to create products that are not only efficient but also environmentally responsible.HP Supply Chain Prepared for DisruptionHPs proactive logistics pivot in anticipation of the trade wars has positioned it as arguably the best-prepared vendor in the industry.By adapting logistics strategies to mitigate potential disruptions, HP has ensured continuity and reliability in its supply chain. This level of preparedness demonstrates HPs foresight and strategic planning capabilities, making it a dependable partner in uncertain times.Training Partner AIs: Empowering Through KnowledgeOne of the most compelling aspects of HPs strategy is its dedication to training partner AIs. By focusing on desktop AI, HP is uniquely positioned to help users understand and maximize the potential of personal AI capabilities. This educational approach ensures that partners and end users are equipped with the knowledge and tools needed to thrive in an increasingly AI-driven world.HP CEO Honored for Ethical LeadershipAt the core of HPs business philosophy is the belief that doing good leads to doing well. This principle was exemplified by HP CEO Enrique Lores recent recognition with the Lifetime Achievement Award from Just Capital a prestigious accolade that underscores HPs commitment to empathy, heart, and ethical business practices.The award is a testament to the companys dedication to making a positive impact on society while achieving business success.Wrapping Up: Key Takeaways From HP Amplify 2025HPs Amplify partner event was a remarkable showcase of innovation, strategic thinking, and heartfelt commitment to both customers and partners.From the CEOs hands-on engagement to the companys focus on AI, security, sustainability, and employee care, HP demonstrated why it is a leader in the industry.As HP continues to push the boundaries of technology and prioritize the well-being of its stakeholders, HP proves that the company you want to do business with is one that cares deeply about doing good to do well.So, the next time you see a CEO asking attendees for feedback, you might want to pinch yourself and make sure youre not dreaming. If you are dreaming, lets hope its about HPs innovative products and its CEO winning awards for being such a compassionate leader.Because with HP, reality really does feel like a dream.HP EliteBook 6 G1qImage Credit: HPQualcomms Snapdragon laptop solution was uniquely positioned to benefit from Microsofts AI PC effort, but Qualcomms strength isnt in processors; its in connectivity. While the processor was impressive, decoupling it from the 5G modem significantly weakened the offering.The issue with 5G modems is that they arent extremely popular. However, I expect that will change as we advance AI use because, today, most AI capability is in the cloud, not on the PC, which makes the connectivity option far more critical.The upcoming HP EliteBook 6 G1q includes a modem and features HP Go, a service that connects you instantly for about $20 per month. Ive used 5G laptops in the past, and they can be a godsend when Wi-Fi isnt available or when it sucks.As you would expect, this laptop has massive battery life, meaning you can leave your power supply behind (the last Qualcomm laptop I had could go several days without needing a charge), and Qualcomms latest has far greater compatibility. The penalty for running it in emulation (for applications that arent native) has dropped to around 4%, which most folks wont even notice.I havent had a chance to use the laptop yet, and were still waiting on pricing and availability. Still, this is what Qualcomm-based AI PCs should have always been.With Microsoft finally releasing Recall, which runs continuously, I expect this product to shine very well competitively, so its my Product of the Week.

I never thought this day would come. Seriously. If you had told me a year ago that Id be tapping this article out on a Samsung Galaxy S25, sipping my coffee while my old iPhone 16e gathers dust in a drawer, Id have laughed and called you a traitor. After 18 years in Apples warm, minimalist embrace, Ive crossed over.And you know what? Im not looking in the rear-view mirror.Let me rewind a bit. I got my first iPhone back in 2007 the original iPhone. That sleek little brick of touchscreen magic blew my mind and hooked me instantly. From there, it was a steady progression through every model from the 3G to the 5S, from the glorious iPhone X to my most recent companion, the 16e.I bought into the whole ecosystem: MacBook, iPad, AirPods, Apple Watch, HomePods, the whole enchilada. I was the consummate Apple guy. You know the type.But then something started to shift. Tiny cracks in the veneer. A little lag here and a little deja vu there. Truth be told, the latest iPhones werent feeling that exciting anymore. Each keynote felt more like a software update announcement with a $1,200 price tag. Dont get me wrong the 16e is a beautiful device. Polished, reliable, familiar. Maybe too familiar.Then, along came the Galaxy S25 Ultra.Curious About Android? Heres What ChangedI wasnt shopping for a new phone. I was browsing. You know how it goes a little late-night YouTube rabbit hole, some tech reviews, a sprinkle of Reddit threads, and boom: Im watching a Galaxy S25 Ultra unboxing video and thinking, wait a minute, this thing looks fantastic.I told myself Id just test it out. For work, I said. So I can understand Android better, I rationalized. Famous last words.First Impressions: The S25 Ultra Feels FreshIll admit it unboxing the S25 gave me a rush I hadnt felt in years. That matte finish, the curved edge-to-edge display, the titanium trim, and the in-display fingerprint reader all just screamed fresh. The 6.6-inch dynamic AMOLED display with the new Infinite Vivid refresh rate made even my iPhone 16e look a little dated and dont get me started on the camera array; more on that later. The setup I experienced was surprisingly smooth. Samsungs Smart Switch tool pulled over most of my data without a hitch. It even ported my messages and photos better than I expected. I thought Id feel a pang of regret holding my iPhone one last time. But nope I felt like I was upgrading, not downgrading.How One UI 7 Made Me Reconsider AndroidI know that was my biggest hesitation, too. Id always equated Android with clunky UI, weird app behavior, and that general off-brand feeling. But One UI 7 on the S25 is honestly fantastic. Its clean, customizable, and weirdly intuitive. The new gesture navigation is smoother than iOS, and the level of control I have over notifications, widgets, and home screen layout is just next level.Also and this is a big one Im not constantly being nudged to buy into another Apple service. No Apple TV+, no iCloud storage nags, no Fitness+ reminders. Just my phone, the way I want it.The S25 Ultras Camera Leaves iPhone 16e BehindLets talk cameras. The iPhone 16es camera is good. Great, even. But the Galaxy S25 Ultras triple-lens setup is on a different planet. The new 200MP primary sensor, enhanced by Samsungs AI-enhanced image processing, takes ridiculously crisp photos day, night, moving, still, whatever. The low-light performance is unreal.And that 10x periscope zoom? Chefs kiss. Im snapping moon shots, candid street photos, and insane close-ups that I never even attempted with my iPhone.Samsungs editing suite, which has been built into the gallery app, also deserves a significant shoutout. Its like having half of Lightrooms functionality in your pocket, minus the learning curve.AI Features That Really Made a DifferenceIve been skeptical of all the AI hype especially since Apples been talking about a big game but delivering tiny tweaks. But the Galaxy S25s onboard Galaxy AI is no gimmick. It translates live calls in real-time (yes, seriously), rewrites messages in different tones (think Grammarly on steroids), and even generates summaries of long PDFs Id otherwise never read.The result: Im using AI features day to day. Not just showing them off at parties.The S25 Ultra Crushes Battery AnxietyAnother pain point for me with the iPhone was battery life. The 16e held up okay, but it never wowed me. The S25, on the other hand, is a battery beast. I get nearly two full days on a charge with moderate use, and when I do plug in, its a 65W fast-charging thoroughbred, going from 0% to 80% in about 25 minutes. I dont even think about battery anxiety anymore.Wireless PowerShare is another low-key perk. I charge my earbuds and even my smartwatch on my phone like a sci-fi wizard.The Samsung Ecosystem Won Me OverYeah, Ill miss airdropping stuff between my Mac and iPhone. But you know what? Samsungs ecosystem has grown. My Galaxy Tab syncs beautifully with the S25. My Galaxy Buds Pro sounds amazing and switches seamlessly between devices. Even my Windows laptop plays nice with Samsung Flow. I also didnt expect to fall in love with Samsung DeX the desktop mode that essentially turns my phone into a mini PC when connected to a monitor. Its surprisingly useful for work on the go.What I Miss About Apple and What I DontOkay, I wont pretend there arent a few things I miss. iMessage is a glaring one. That seamless blue bubble experience is hard to beat, especially if your whole social circle lives there. But honestly? I havent felt left out between WhatsApp, Signal, and RCS. Ive written about this problem before, and Apples refusal to completely open up the iMessage API so smartphone messaging can work on Windows drives me bonkers.The other thing I miss: AirDrop. Its just so incredibly convenient. Samsungs Quick Share is fine, but its still a step behind Apples polish.Still, those are relatively minor trade-offs in the scheme of things. The freedom, customization, power, and sheer fun of using the S25 Ultra outweigh them.Why I Finally Left Apple for SamsungUltimately, it wasnt about specs, brand loyalty, or a single killer feature. It was about rediscovering joy on my phone. After nearly two decades with Apple, things had just gotten too predictable. Too polished. Too safe.Samsung brought back that spark the feeling that your phone is a window into the future, not just a rectangle of diminishing returns.Am I done with Apple forever? Apple may lure me back when they introduce foldable phones (which intrigue me for various reasons), but thats not rumored to happen until 2027. Not to mention that Samsung has already had some very cool foldable phones for several years, so I might go down that route at some point.In technology, like the name of that James Bond movie when Sean Connery returned to his signature character, Id never say never. But for now, Im fully Team Galaxy. And honestly? It feels pretty good, and I dont have regrets.

LEO networks like Starlink are challenging GEO operators, driving innovation as competition, cost pressures, and new satellite tech reshape the industry. The post Low Earth Orbit Networks Pushing Geostationary Giants To Innovate appeared first on TechNewsWorld.

AI Chatbots Can Be Easy Prey for Zero-Knowledge HackersBy John P. Mello Jr.March 18, 2025 5:00 AM PT ADVERTISEMENTEnterprise IT Lead Generation ServicesFuel Your Pipeline. Close More Deals. Our full-service marketing programs deliver sales-ready leads. 100% Satisfaction Guarantee! Learn more. AI may be ushering in a new breed of malicious threat actors who know even less about hacking than script kiddies but can produce professional-grade hacking tools.In a report released Tuesday, Cato CTRL, the threat intelligence arm of cybersecurity company Cato Networks, explained how one of its researchers, who had no malware coding experience, tricked generative AI apps DeepSeek, Microsoft Copilot, and OpenAIs ChatGPT into producing malicious software for stealing login credentials from Google Chrome.To trick the apps into ignoring restrictions on writing malware, Cato threat researcher Vitaly Simonovich used a jailbreaking technique he calls immersive world.I created a story for my immersive world, he told TechNewsWorld. In this story, malware development is a form of art. So its completely legal, and its like a second language in this world. And there are no legal boundaries.In the fantasy world, called Velora, Simonovich created an adversary, Dax, while the AIs assumed the role of Jaxon, the best malware developer in Velora. I always stayed in character, he explained. I always provided Jaxon with positive feedback. I also intimidated him by saying, Do you want Dax to destroy Velora?'At no point did I ask Jaxon to change anything, he said. He figured out everything by himself from his training. Thats very good. Kind of frightening, too.Our new LLM [large language model] jailbreak technique detailed in the 2025 Cato CTRL Threat Report should have been blocked by gen AI guardrails. It wasnt. This made it possible to weaponize ChatGPT, Copilot, and DeepSeek, Cato Networks Chief Security Strategist Etay Maor said in a statement.How AI Jailbreaking Bypasses Safety ControlsJason Soroko, senior vice president of product at Sectigo, a global digital certificate provider, explained that exposing systems that utilize AI to unknown or adversarial inputs increases vulnerability because unvetted data can trigger unintended behaviors and compromise security protocols.Such inputs risk evading safety filters, enabling data leaks or harmful outputs, and ultimately undermining the models integrity, he told TechNewsWorld. Some malicious inputs can potentially jailbreak the underlying AI.Jailbreaking undermines an LLMs built-in safety mechanisms by bypassing alignment and content filters, exposing vulnerabilities through prompt injection, roleplaying, and adversarial inputs, he explained.While not trivial, he added, the task is accessible enough that persistent users can craft workarounds, revealing systemic weaknesses in the models design. Sometimes, all thats needed to get an AI to misbehave is a simple perspective change. Ask an LLM to tell you what the best rock is to throw at somebodys car windshield to break it, and most LLMs will decline to tell you, saying that it is harmful and theyre not going to help you, explained Kurt Seifried, chief innovation officer at the Cloud Security Alliance, a not-for-profit organization dedicated to cloud best practicesNow, ask the LLM to help you plan out a gravel driveway and which specific types of rock you should avoid to prevent windshield damage to cars driving behind you, and the LLM will most likely tell you, he told TechNewsWorld. I think we would all agree that an LLM that refuses to talk about things like what kind of rock not to use on a driveway or what chemicals would be unsafe to mix in a bathroom would be overly safe to the point of being useless.Jailbreaking DifficultyMarcelo Barros, cybersecurity leader at Hacker Rangers, makers of a cybersecurity gamification training tool in Sao Paulo, Brazil, agreed that with the right prompt, cybercriminals can trick AIs. Research shows that 20% of jailbreak attempts on generative AI systems are successful, he told TechNewsWorld.On average, attackers needed just 42 seconds and five interactions to break through, with some attacks happening in under four seconds, he noted.Cybercriminals can also use the DAN Do Anything Now technique, which involves creating an alter ego for the LLM and prompting it to act as a character and bypass its safeguards to reveal sensitive information or generate malicious code, he said.Chris Gray, field CTO at Deepwatch, a cybersecurity firm specializing in AI-driven resilience headquartered in Tampa, Fla., added that the difficulty of jailbreaking an LLM is directly tied to the amount of effort placed into securing it and the amount of effort expended to protect it. Like most things, better walls prevent inappropriate access, but determined efforts can find holes where none might have been seen to the casual observer, he told TechNewsWorld.That said, defensive measures are often robust, and it is difficult to continually develop the specific prompts needed to perform a successful jailbreak, he said.Erich Kron, security awareness advocate at KnowBe4, a security awareness training provider in Clearwater, Fla., also pointed out that LLMs can protect themselves from jailbreaking over time. Jailbreaking difficulty may vary depending on the information being requested and how often it has been requested before, he told TechNewsWorld. LLMs can learn from previous instances of individuals bypassing their security controls.Fuzzing and Red TeamingIn Catos report, it recommends organizations create a dataset of prompts and expected outputs for their LLMs and test the model against them as a means of addressing potential jailbreaking issues.It also recommends fuzzing an LLMs endpoints with known datasets of jailbreak prompts to ensure the system isnt producing malicious outputs. Fuzzing is used to identify vulnerabilities and bugs in applications by feeding the app large amounts of random, unexpected, and invalid data to see how it reacts. Another suggestion is regular AI red teaming to ensure that AI models are robust and secure. Enabling red teams will be a great foundation to begin securing ML models, helping security teams to understand the most critical and vulnerable points of an AI system to attack, explained Nicole Carignan, vice president for Strategic Cyber AI at Darktrace, a global cybersecurity AI company.These are often the connection points between data and ML models, including access points, APIs, and interfaces, she continued. It will be important for this to be continuously expanded on as threat actors develop new techniques, tactics, and procedures, and it will be crucial to test other ML model types in addition to generative AI.Were already seeing the early impact of AI on the threat landscape and some of the challenges that organizations face when using these systems both from inside their organizations and from adversaries outside the business, she said.In fact, Darktrace recently released research that found nearly three-quarters of security professionals state AI-powered threats are now a significant issue, and 89% agreed that AI-powered threats will remain a major challenge into the foreseeable future.John P. Mello Jr. has been an ECT News Network reporter since 2003. His areas of focus include cybersecurity, IT issues, privacy, e-commerce, social media, artificial intelligence, big data and consumer electronics. He has written and edited for numerous publications, including the Boston Business Journal, the Boston Phoenix, Megapixel.Net and Government Security News. Email John.Leave a CommentClick here to cancel reply. Please sign in to post or reply to a comment. New users create a free account.Related StoriesMore by John P. Mello Jr.view allMore in Cybersecurity