The European Union (EU) AI Act may seem like a done deal, but stakeholders are still drafting the code of practice that will lay out rules for general-purpose AI (GPAI) models, including those with systemic risk. Now, though, as that drafting process approaches its deadline, US President Donald Trump is reportedly pressuring European regulators to scrap the rulebook. The US administration and other critics claim that it stifles innovation, is burdensome, and extends the bounds of the AI law, essentially creating new, unnecessary rules. The US government’s Mission to the EU recently reached out to the European Commission and several European governments to oppose its adoption in its current form, Bloomberg reports. “Big tech, and now government officials, argue that the draft AI rulebook layers on extra obligations, including third party model testing and full training data disclosure, that go beyond what is in the legally binding AI Act’s text, and furthermore, would be very challenging to implement at scale,” explained Thomas Randall, director of AI market research at Info-Tech Research Group. Onus is shifting from vendor to enterprise On its web page describing the initiative, the European Commission said, “the code should represent a central tool for providers to demonstrate compliance with the AI Act, incorporating state-of-the-art practices.” The code is voluntary, but the goal is to help providers prepare to satisfy the EU AI Act’s regulations around transparency, copyright, and risk mitigation. It is being drafted by a diverse group of general-purpose AI model providers, industry organizations, copyright holders, civil society representatives, members of academia, and independent experts, overseen by the European AI Office. The deadline for its completion is the end of April. The final version is set to be presented to EU representatives for approval in May, and will go into effect in August, one year after the AI Act came into force. It will have teeth; Randall pointed out that non-compliance could draw fines of up to 7% of global revenue, or heavier scrutiny by regulators, once it takes effect. But whether or not Brussels, the de facto capital of the EU, relaxes or enforces the current draft, the weight of ‘responsible AI’ is already shifting from vendors to the customer organizations deploying the technology, he noted. “Any organization conducting business in Europe needs to have its own AI risk playbooks, including privacy impact checks, provenance logs, or red-team testing, to avoid contractual, regulatory, and reputational damages,” Randall advised. He added that if Brussels did water down its AI code, it wouldn’t just be handing companies a free pass, “it would be handing over the steering wheel.” Clear, well-defined rules can at least mark where the guardrails sit, he noted. Strip those out, and every firm, from a garage startup to a global enterprise, will have to chart its own course on privacy, copyright, and model safety. While some will race ahead, others will likely have to tap the brakes because the liability would “sit squarely on their desks.” “Either way, CIOs need to treat responsible AI controls as core infrastructure, not a side project,” said Randall. A lighter touch regulatory landscape If other countries were to follow the current US administration’s approach to AI legislation, the result would likely be a lighter touch regulatory landscape with reduced federal oversight, noted Bill Wong, AI research fellow at Info-Tech Research Group. He pointed out that in January, the US administration issued Executive Order 14179, “Removing Barriers to American Leadership in Artificial Intelligence.” Right after that, the National Institute of Standards and Technology (NIST) updated its guidance for scientists working with the US Artificial Intelligence Safety Institute (AISI). Further, references to “AI safety,” “responsible AI,” and “AI fairness” were removed; instead, a new emphasis was placed on “reducing ideological bias to enable human flourishing and economic competitiveness.” Wong said: “In effect, the updated guidance appears to encourage partners to align with the executive order’s deregulatory stance.”

As AI integration accelerates, businesses are facing widening skills gaps that traditional employment models struggle to address, and so more companies are choosing to hire freelancers to fill the void, according to a new report. The report, from freelance work platform Upwork, claims there’s a major workforce shift as well, with 28% of skilled knowledge workers now freelancing for greater “autonomy and purpose.” An additional 36% of full-time employees are considering switching to freelance, while only 10% of freelancers want to return to traditional jobs, according to the report, which is based on a recent survey of 3,000 skilled, US-based knowledge workers. Gen Z is leading the shift, as 53% of skilled Gen Z professionals already freelance, and they’re expected to make up 30% of the freelance US workforce by 2030. “The traditional 9-to-5 model is rapidly losing its grip as skilled talent chooses flexibility, financial control, and meaningful work over outdated corporate structures,” Kelly Monahan, managing director of the Upwork Research Institute, said in a statement. “Companies that cling to old hiring and workforce models risk falling behind.” US businesses have ramped up their freelance hiring by 260% between 2022 and 2024, according to a report from Mellow.io, an HR platform that manages and pays freelance contractors. Increasingly, US businesses have turned to freelancers overseas — most frequently Eastern Europe — to fill their tech talent void, particularly for web developers, programmers, data analysts, and web designers, according to Mellow.io. “This trend shows no signs of slowing,” Mellow.io’s report stated. “The region offers an unparalleled balance of cost efficiency and highly skilled talent.” The US is a freelancer haven Gig workers, earning through short-term, flexible jobs via apps or platforms, are also thriving, according to career site JobLeads. JobLeads analyzed data from the Online Labour Observatory and the World Bank Group to reveal the countries dominating online gig work. The United States is leading in the number of online freelancers with 28% of the global online freelance market. Software and tech roles dominate in the US, representing 36.4% of freelancers, followed by creative/multimedia (21.1%) and clerical/data entry jobs (18.2%). Globally, Spain and Mexico rank second and third in freelancer share, with 7.0% and 4.6%, respectively. Among full-time online gig workers, 52% have a high school diploma, while 20% hold a bachelor’s degree, according to JobLeads. “The gig economy is booming worldwide, with the number of gig workers expected to rise by over 30 million in the next year alone,” said Martin Schmidt, JobLead’s managing director. “This rapid growth reflects a fundamental shift in how people approach work — flexibility and autonomy are no longer just perks but non-negotiables for today’s workforce.” Gen Z and younger professionals are embracing gig work for its flexibility and control, while businesses gain access to a global pool of skilled freelancers, Schmidt said. “As the sector continues to evolve, both workers and employers need to adapt to a new reality where traditional employment models may no longer meet the needs and expectations of the modern workforce,” he said. Confidence in freelancing is high, with 84% of freelancers and 77% of full-time workers viewing its future as bright, according to Upwork’s report. Freelancers are also seeing more opportunities, with 82% reporting more work than last year, compared to 63% of full-time employees, the report said. Freelance workers generated $1.5 trillion in earnings in 2024 alone, Upwork said. The trend is gaining momentum, particularly among Gen Z, with many full-time employees eyeing independent work, according to Upwork. Freelancers are leading in AI, software, and sustainability jobs, demonstrating higher adaptability and continuous learning, according to the report, which focused exclusively on skilled knowledge workers, not gig workers. It also included “moonlighters,” or workers who have full-time employment but freelance on the side. More than half (54%) of freelancers report advanced AI proficiency compared to 38% of full-time employees, and 29% have extensive experience building, training, and fine-tuning machine learning models (vs. 18% of full-time employees), the report stated. Those who earn exclusively through freelance work report a median income of $85,000, surpassing their full-time employee counterparts at $80,000, the report stated. Upwork’s Future Workforce Index is the company’s first such report, and so it said it is unable to provide freelance employment numbers from previous years that would indicate a rising or falling trend. “However, what we can confidently say, based on multiple studies conducted by the Upwork Research Institute over the past several years, is that freelancing isn’t a passing trend,” an Upwork spokesperson said. “It continues to hold steady and accelerate, emerging as a vital and intentional component of the skilled workforce.” A silver tsunami Emily Rose McRae, a Gartner Resarch senior director analyst, said she’s seeing growing interest in freelancing from professionals who desire more flexible work, oftentimes as a safety net for people who lost their job during economic turmoil “and also as a way to build up a network of clients when starting a new business or looking to expand your small business.” Organizations are also facing an impending “silver tsunami” of older workers retiring and leaving a talent gap in their wake. “Many clients I speak with on this topic are trying to identify the best strategy for addressing this expertise gap, whether it is upskilling more junior employees, bringing retired experts back to serve as freelance mentors or coaches, contracting out critical projects to experts on a freelance market, or even redesigning roles and workflows to reduce the amount of expertise needed,” she said. “Being able to bring past employees back as freelancers can be critical for knowledge management and training,” McRae said. “This is especially critical when increasingly AI tools are being deployed on the basic and repetitive tasks that were previously the training ground used for employees to create a pipeline of future experts within the employee base,” McRae said. Despite the occurrence of layoffs — and sometimes because of them — organizations often face skills gaps exacerbated by the rise of AI, according to Forrester Research. Skills intelligence tools, often powered by AI, can help organizations identify and manage the skills and gaps in their workforce and predict future skill needs, including recommending needed recruiting, upskilling and reskilling, and talent mobility. Companies must also be able to scale up or down rapidly in on-demand talent markets, which include contractors, freelancers, gig workers, and service providers. On-demand talent increases the adaptability of your workforce but works best for non-core functions and for specialized skills that are needed for a limited period. Companies, however, can’t simply replace employees with freelancers without facing significant risks, McRae noted. Freelancers are best used for defined projects with clear deliverables. Using them to do the same work as former employees, without changing the role or workflow, can lead to legal and operational issues, she said. As reliance on non-employees grows, so do risks like worker misclassification, dual employment, compliance problems, and costly mistakes such as rehiring underperforming contractors or overpaying for services. “I’ll see this at organizations that instituted hiring freezes, so business leaders turned to contractors to continue to be able to meet their goals,” she said. “It can also create financial risks — when there isn’t much transparency or data collection going on, organizations may find that they are paying the same contractor service provider or freelancer different rates in different departments, for the same set of tasks.” There’s also a risk that third-party contractors are not vetting temp workers, who may not meet the necessary certifications and trainings to comply with local or national regulations, McRae added. “Or that contractors and freelancers have not been fully offboarded after completing their assignments and still retain access to the organization’s systems and data,” she said.

Apple is on track to source all the iPhones it sells in the US from India by the end of next year as politically driven tensions drive a wedge between the US and China. This is a major move that the company has been building toward for some time, but the recent tariffs announcements may have accelerated the plan. Designed by Apple in California, Made in India If you’ve been following my work you’ll already be aware of the importance India now has for Apple. India is expected to be the manufacturing hub for 25% of all the iPhones sold globally by the end of the year, and now the Financial Times reports that Apple aims to make 60 million iPhones in India “as soon as next year” — though other sources suggest that target may prove too ambitious. To support this transition, Apple and its partners have invested billions in building their businesses there. Foxconn is currently building its second-largest factory outside China in India. At a cost of $2.5 billion, the facility will create 40,000 jobs and double its manufacturing capacity in India. India’s biggest conglomerate, the Tata Group, also makes iPhones for Apple using facilities formerly owned by Pegatron and Wistron. To support the project, Apple is also encouraging component manufacturers to set up shop in India, with India’s government recently announcing a range of incentives to help encourage them to do so. The idea behind this is, of course, to ensure that the iPhones assembled in India make use of components that are also manufactured there in order to minimize the cost of any tariffs. These investments are accompanied by a range of external improvements, including improved infrastructure. While there have been no major signals to this effect as yet, it is becoming increasingly likely that Apple will eventually commence manufacturing other products in India at some point. This didn’t — and couldn’t — happen overnight What’s important to note is that none of this is happening suddenly. This has already taken years. Apple has been working on its journey to India for almost a decade, presumably since before Apple CEO Tim Cook made his first disclosed visit to the nation. During that visit, Cook stressed that his company intended taking position in India, saying: “We’re not here for a quarter, or two quarters, or the next year, or the next year. We’re here for a thousand years.” Apple had originally intended simply to set up retail stores there and build a business from India’s burgeoning middle class, but very swiftly saw the sense of transitioning some production to the nation, accelerating these plans once the Covid plague threw international supply chains into chaos. In other words, while the company’s move to make iPhones for US market in India may seem sudden, it is something that has taken years. That effort proves that shifting manufacturing to new nations is not an overnight task; it takes time, a lot of investment, an available and accessible skilled workforce, and more. With that in mind, it is foolish to expect manufacturing infrastructure to migrate across territorial boundaries any faster than Apple — with all its advantages — has been able to achieve in India. The overall impact of Apple’s more diversified approach is a decreased reliance on its former manufacturing partner, China, from which India and other nations, including Brazil, Vietnam, and Thailand, are seeing some benefit as Apple is also increasing its manufacturing capabilities in those countries. Doing the business The company’s moves into India aren’t just about tariff avoidance. The decision to base more manufacturing there has helped Apple capture hearts and minds among consumers there, translating into accelerating business results. Driven by the iPhone 16 series, Apple sold three million iPhones for the first time in India in Q1 2025, according to IDC, its largest-ever shipment in the nation. “Apple achieved its best-ever Q1 in India, driven by strong iPhone 16 series momentum,” said Canalys. In other words, it looks as if Apple’s shrewd decision to create business in India will pay a double benefit, giving the company access to a growing and developing economy even as the so-called ‘First World’ economies tumble into existential decline. You can follow me on social media! Join me on BlueSky,  LinkedIn, Mastodon, and MeWe. 

We’re all talking to fake people now, but most people don’t realize that interacting with AI is a subtle and powerful skill that can and should be learned. The first step in developing this skill set is to acknowledge to yourself what kind of AI you’re talking to and why you’re talking to it. AI voice interfaces are powerful because our brains are hardwired for human speech. Even babies’ brains are tuned to voices before they can talk, picking up language patterns early on. This built-in conversational skill helped our ancestors survive and connect, making language one of our most essential and deeply rooted abilities. But that doesn’t mean we can’t think more clearly about how to talk when we speak to AI. After all, we already speak differently to other people in different situations. For example, we talk one way to our colleagues at work and a different way to our spouses. Yet people still talk to AI like it’s a person, which it’s not; like it can understand, which it cannot; and like it has feelings, pride, or the ability to take offense, which it doesn’t. The two main categories of talking AI It’s helpful to break the world of talking AI (both spoken and written) into two categories: Fantasy role playing, which we use for entertainment. Tools, which we use for some productive end, either to learn information or to get a service to do something useful for us. Let’s start with role-playing AI. AI for pretending You may have heard of a site and app called Status AI, which is often described as a social network where everyone else on the network is an AI agent. A better way to think about it is that it’s a fantasy role-playing game in which the user can pretend to be a popular online influencer. Status AI is a virtual world that simulates social media platforms. Launched as a digital playground, it lets people create online personas and join fan communities built around shared interests. It “feels” like a social network, but every interaction—likes, replies, even heated debates—comes from artificial intelligence programmed to act like real users, celebrities, or fictional characters. It’s a place to experiment, see how it feels to be someone else, and interact with digital versions of celebrities in ways that aren’t possible on real social media. The feedback is instant, the engagement is constant, and the experience, though fake, is basically a game rather than a social network. Another basket of role-playing AI comes from Meta, which has launched AI-powered accounts on Facebook, Instagram, and WhatsApp that let users interact with digital personas — some based on real celebrities like Tom Brady and Paris Hilton, others entirely fictional. These AI accounts are clearly labeled as such, but (thanks to AI) can chat, post, and respond like real people. Meta also offers tools for influencers to use AI agents to reply to fans and manage posts, mimicking their style. These features are live in the US, with plans to expand, and are part of Meta’s push to automate and personalize social media. Because these tools aim to provide make-believe engagements, it’s reasonable for users to pretend like they’re interacting with real people. These Meta tools attempt to cash in on the wider and older phenomenon of virtual online influencers. These are digital characters created by companies or artists, but they have social media accounts and appear to post just like any influencer. The best-known example is Lil Miquela, launched in 2016 by the Los Angeles startup Brud, which has amassed 2.5 million Instagram followers. Another is Shudu, created in 2017 by British photographer Cameron-James Wilson, presented as the world’s first digital supermodel. These characters often partner with big brands. A post by one of the major virtual influencer accounts can get hundreds or thousands of likes and comments. The content of these comments ranges from admiration for their style and beauty to debates about their digital nature. Presumably, many people think they’re commenting to real people, but most probably engage with a role-playing mindset. By 2023, there were hundreds of these virtual influencers worldwide, including Imma from Japan and Noonoouri from Germany. They’re especially popular in fashion and beauty, but some, like FN Meka, have even released music. The trend is growing fast, with the global virtual influencer market estimated at over $4 billion by 2024. AI for knowledge and productivity We’re all familiar with LLM-based chatbots like ChatGPT, Gemini, Claude, Copilot, Meta AI, Mistral, and Perplexity. The public may be even more familiar with non-LLM assistants like Siri, Google Assistant, Alexa, Bixby, and Cortana, which have been around much longer. I’ve noticed that most people make two general mistakes when interacting with these chatbots or assistants. The first is that they interact with them as if they’re people (or role-playing bots). And the second is that they don’t use special tactics to get better answers. People often treat AI chatbots like humans, adding “please,” “thank you,” and even apologies. But the AI doesn’t care, remember, and is not significantly affected by these niceties. Some people even say “hi” or “how are you?” before asking their real questions. They also sometimes ask for permission, like “Can you tell me…” or “Would you mind…” which adds no value. Some even sign off with “goodbye” or “thanks for your help,” but the AI doesn’t notice or care. Politeness to AI wastes time — and money! A year ago, Wharton professor Ethan Mollick pointed out that people using “please” and “thank you” in AI prompts add extra tokens, which increases the compute power needed by the LLM chatbot companies. This concept resurfaced on April 16 of this year, when OpenAI CEO Sam Altman replied to another user on X, saying (perhaps exaggerating) that polite words in prompts have cost OpenAI “tens of millions of dollars.” “But wait a second, Mike,” you say. “I heard that saying ‘please’ to AI chatbots gets you better results.” And that’s true — sort of. Several studies and user experiments have found that AI chatbots can give more helpful, detailed answers when users phrase requests politely or add “please” and “thank you.” This happens because the AI models, trained on vast amounts of human conversation, tend to interpret polite language as a cue for more thoughtful responses. But prompt engineering experts say that clear, specific prompts — such as giving context or stating exactly what you want — consistently produce much better results than politeness. In other words, politeness is a tactic for people who aren’t very good at prompting AI chatbots. The best way to get top-quality answers from AI chatbots is to be specific and direct in your request. Always say exactly what you want, using clear details and context. Another powerful tactic is something called “role prompting” — tell the chatbot to act as a world-class expert, such as, “You are a leading cybersecurity analyst,” before asking a question about cybersecurity. This method, proven in studies like Sander Schulhoff’s 2025 review of over 1,500 prompt engineering papers, leads to more accurate and relevant answers because it tells the chatbot to favor content in the training data produced by experts, rather than just lumping the expert opinion in with the uneducated viewpoints. Also: Give background if it matters, like the audience or purpose. (And don’t forget to fact-check responses. AI chatbots often lie and hallucinate.) It’s time to up your AI chatbot game. Unless you’re into using AI for fantasy role playing, stop being polite. Instead, use prompt engineering best practices for better results. >We’re all talking to fakepeople now, but most people don’t realize that interacting with AI is a subtleand powerful skill that can and should be learned.The first step in developingthis skill set is to acknowledge to yourself what kind of AI you’re talking toand why you’re talking to it. AI voice interfaces arepowerful because our brains are hardwired for human speech. Even babies’ brainsare tuned to voices before they can talk, picking up language patterns earlyon. This built-in conversational skill helped our ancestors survive and connect,making language one of our most essential and deeply rooted abilities.But that doesn’t mean we can’tthink more clearly about how to talk when we speak to AI. After all, we alreadyspeak differently to other people in different situations. For example, we talkone way to our colleagues at work and a different way to our spouses. Yet people still talk to AIlike it’s a person, which it’s not; like it can understand, which it cannot;and like it has feelings, pride, or the ability to take offense, which itdoesn’t. The two main categories oftalking AIIt’s helpful to break theworld of talking AI (both spoken and written) into two categories: >1.    Fantasy role playing, which we use forentertainment. >2.    Tools, which we use for some productive end,either to learn information or to get a service to do something useful for us. Let’s start with role-playingAI. id="ai-for-pretending">AI for pretending>You may have heard of a siteand app called Status AI, which is oftendescribed as a social network where everyone else on the network is an AIagent. A better way to think about itis that it’s a fantasy role-playing game in which the user can pretend to be apopular online >influencer. Status AI is a virtual worldthat simulates social media platforms. Launched as a digital playground, itlets people create online personas and join fan communities built around sharedinterests. It “feels” like a social network, but every interaction—likes,replies, even heated debates—comes from artificial intelligence programmed toact like real users, celebrities, or fictional characters.It’s a place to experiment,see how it feels to be someone else, and interact with digital versions ofcelebrities in ways that aren’t possible on real social media. The feedback isinstant, the engagement is constant, and the experience, though fake, isbasically a game rather than a social network. Another basket of role-playingAI comes from Meta, which has launchedAI-powered accounts on Facebook, Instagram, and WhatsApp that let usersinteract with digital personas — some based on real celebrities like Tom Bradyand Paris Hilton, others entirely fictional. These AI accounts are clearlylabeled as such, but (thanks to AI) can chat, post, and respond like realpeople. Meta also offers tools for influencers to use AI agents to reply tofans and manage posts, mimicking their style. These features are live in theUS, with plans to expand, and are part of Meta’s push to automate andpersonalize social media.Because these tools aim toprovide make-believe engagements, it’s reasonable for users to pretend likethey’re interacting with real people. These Meta tools attempt tocash in on the wider and older phenomenon of virtual online influencers. Theseare digital characters created by companies or artists, but they have socialmedia accounts and appear to post just like any influencer. The best-knownexample is Lil Miquela, launched in 2016 by the Los Angeles startup Brud, whichhas amassed 2.5 million Instagram followers. Another is Shudu, created in 2017by British photographer Cameron-James Wilson, presented as the world’s firstdigital supermodel. These characters often partner with big brands. A post by one of the majorvirtual influencer accounts can get hundreds or thousands of likes andcomments. The content of these comments ranges from admiration for their styleand beauty to debates about their digital nature. Presumably, many people thinkthey’re commenting to real people, but most probably engage with a role-playingmindset. By 2023, there were hundredsof these virtual influencers worldwide, including Imma from Japan and Noonoourifrom Germany. They’re especially popular in fashion and beauty, but some, likeFN Meka, have even released music. The trend is growing fast, with the globalvirtual influencer market estimated at over $4 billion by 2024. id="ai-for-knowledge-and-productivity">AI for knowledge and productivity>We’re all familiar withLLM-based chatbots like ChatGPT, Gemini, Claude, Copilot, Meta AI, Mistral, andPerplexity. The public may be even morefamiliar with non-LLM assistants like Siri, Google Assistant, Alexa, Bixby, andCortana, which have been around much longer.I’ve noticed that most peoplemake two general mistakes when interacting with these chatbots or assistants.The first is that theyinteract with them as if they’re people (or role-playing bots). And the secondis that they don’t use special tactics to get better answers. People often treat AI chatbotslike humans, adding “please,” “thank you,” and evenapologies. But the AI doesn’t care, remember, and is not significantly affectedby these niceties. Some people even say “hi” or “how areyou?” before asking their real questions. They also sometimes ask forpermission, like “Can you tell me…” or “Would you mind…”which adds no value. Some even sign off with “goodbye” or“thanks for your help,” but the AI doesn’t notice or care. Politeness to AI wastes time —and money! A year ago, Wharton professor Ethan Mollick pointed out that peopleusing “please” and “thank you” in AI prompts add extratokens, which increases the compute power needed by the LLM chatbot companies.This concept resurfaced on April 16 of this year, when OpenAI CEO Sam Altmanreplied to anotheruser on X, confirming that polite words in prompts have cost OpenAI “tens of millions ofdollars.” “But wait a second,Mike,” you say. “I heard that saying ‘please’ to AI chatbots gets youbetter results.” And that’s true — sort of. Several studies and userexperiments have found that AI chatbots can give more helpful, detailed answerswhen users phrase requests politely or add “please” and “thankyou.” This happens because the AI models, trained on vast amounts of humanconversation, tend to interpret polite language as a cue for more thoughtfulresponses.But prompt engineering expertssay that clear, specific prompts — such as giving context or stating exactlywhat you want — consistently produce much better results than politeness. In other words, politeness isa tactic for people who aren’t very good at prompting AI chatbots. The best way to gettop-quality answers from AI chatbots is to be specific and direct in yourrequest. Always say exactly what you want, using clear details and context. Another powerful tactic issomething called “role prompting” — tell the chatbot to act as aworld-class expert, such as, “You are a leading cybersecurityanalyst,” before asking a question about cybersecurity. This method,proven in studies like SanderSchulhoff’s 2025 review of over 1,500 prompt engineering papers, leads tomore accurate and relevant answers because it tells the chatbot to favorcontent in the training data produced by experts, rather than just lumping theexpert opinion in with the uneducated viewpoints. Also: Give background if itmatters, like the audience or purpose. (And don’t forget tofact-check responses. AI chatbots often lie and hallucinate.)It’s time to up your AI chatbot game.Unless you’re into using AI for fantasy role playing, stop being polite.Instead, use prompt engineering best practices for better results.

Psst: Come close. Your Android phone has a little-known superpower — a futuristic system for bridging the physical world around you and the digital universe on your device. It’s one of Google’s best-kept secrets. And it can save you tons of time and effort. Oh — and no, it isn’t Gemini. It’s a little somethin’ called Google Lens, and it’s been lurking around on Android and quietly getting more and more capable for years — since long before “AI” became part of our popular vernacular. Google doesn’t make a big deal about it, weirdly enough, and you really have to go out of your way to even realize it exists. But once you uncover it, you’ll feel like you have a magic wand in your pocket. At its core, Google Lens is best described as a search engine for the real world. It uses (yes…) artificial intelligence to identify text and objects both within images and in a live view from your phone’s camera, and it then lets you learn about and interact with those elements in all sorts of interesting ways. But while Lens’s ability to, say, identify a flower, look up a book, or give you info about a landmark is certainly impressive, it’s the system’s more mundane-seeming productivity powers that are far more likely to find a place in your day-to-day life. So grab your nearest Android gadget, go install the Google Lens app, if you haven’t already — or take your pick from any of the other smart Google-Lens-launching shortcuts — and get ready to teach your phone some spectacularly useful new tricks. [Hey — love shortcuts? My free Android Shortcut Supercourse will teach you tons of time-saving tricks for your phone. Sign up now and start learning!] Google Lens trick #1: Dive deep into your screen >In a mildly wild twist, the first and newest Google Lens goody in our list is also the oldest and most familiar one of all — at least, if you’ve paying attention in this arena for long. >It’s a snazzy new feature that lets you indirectly have Lens analyze whatever’s on your screen and then give you helpful extra context around it. >This one can actually be accessed via Google’s next-gen Gemini virtual assistant. Just summon Gemini, using the “Hey Google” hotword or any other method you like, then look for the tappable “Ask about screen” button within its overlay interface. The “Ask about screen” button within Gemini is a hidden way to access a powerful Lens feature.JR Raphael, Foundry While the answer is wrapped in Gemini, the technology powering it is the same stuff that’s been present within Lens for ages. And it’s every bit as impressive. Answers, on demand — from anywhere on Android.JR Raphael, Foundry >And if you’re feeling a pesky sense of déjà vu around this, well, you should be: Google first announced this latest iteration of the on-demand screen searching system more than two years ago, for the previous-gen Google Assistant system. Prior to that point, Assistant had briefly offered a similar sort of setup >without> Lens’s involvement. And prior to that, Google had a spectacularly useful native Android feature called Now on Tap, way back in 2015’s Android 6.0 (Marshmallow) era — though amusingly, we haven’t >quite> caught back up to that level of search intelligence just yet. >Hey, what can we say? It’s the Google way. Google Lens trick #2: Copy text from the real world From the virtual world to the physical world around us, Google Lens’s most potent power and the one I rely on most frequently is its ability to grab text from a physical document — a paper, a book, a whiteboard, a suspiciously wordy tattoo on your rumpus, or anything else with writing on it — and then copy that text onto your phone’s clipboard. From there, you can easily paste the text into a Google Doc, a note, an email, a Slack chat, or anywhere else imaginable. To do that, just open up Google Lens, point your device’s camera at any text around you, then tap the big circular search icon — and you’ll be able to use your finger to select the exact portion of text you want as if it were regular ol’ digital text on a website. All that’s left is to hit the “Copy” command in the pop-up alongside it, and every last word will be on your system clipboard and ready to paste wherever your thumpy little heart desires. You can copy text from anywhere — virtual or physical — with a little help from Lens.JR Raphael, Foundry Google Lens trick #3: Connect text to your computer Let’s face it: Most of us aren’t working only from our Android phones. If you need to get some real-world text onto your computer, Lens can handle that for you, too. Just go through the same steps we did a second ago, but this time, look for the “Copy to computer” option in the same pop-up menu. (You might have to tap a three-dot icon within that pop-up to reveal it.) As long as you’re actively signed into Chrome with the same Google account on a computer — any computer, whether it’s Windows, Mac, Linux, or ChromeOS — that option should appear. And when you tap it, you’ll get a list of all available destinations. Get any text onto your computer’s clipboard in an instant with Lens’s clever copying commands.JR Raphael, Foundry Pick the device you want, and just like magic, the text from the physical document will be on that computer’s clipboard — ready and waiting to be pasted wherever you want it. Hit Ctrl-V (or Cmd-V, on a Mac), and shazam! It’ll pop into any text field, in any app or process where pasting is supported. Google Lens trick #4: Hear anything out loud Maybe you’ve just been handed a long memo, a printed-out brief of some sort, or a letter from your dear Aunt Sally. Whatever it is, give your eyes a breather and let Lens read it for you while you’re on the go and between meetings. Just point your phone at the paper, exactly as we did before, and select some specific text within the image once more. This time, look for the little “Listen” option in the pop-up panel atop your image. Pound your pinky down on that bad boy, and the Google Lens app will actually read the selected text out loud to you, in a soothingly pleasant voice. Hey, Google: How ’bout a nap-time story while we’re at it?! Google Lens trick #5: Ask, ask, ask away In a move that now seems foreshadowing, Lens has the completely concealed ability to let you chat out loud and ask anything imaginable about whatever your device’s camera is showing. All you’ve gotta do to try it is open up Lens, aim your camera at something, and then press and hold Lens’s big search button. Then, you can simply speak aloud and ask anything on your mind in a completely natural, conversational way. Pressing and holding the Lens search button lets you talk and ask questions in a completely natural way.JR Raphael, Foundry Google’s official introduction of the feature involved asking questions about why some sort of product isn’t working as expecting or how you can fix some common real-world maintenance issue, but it can be every bit as helpful for practically any purpose — anytime you find yourself facing a question about something in front of you. This one, for now, seems to be available only within the U.S. and in English. Google Lens trick #6: Interact with text from an image In addition to the live stuff, Lens can pull and process text from images — including both actual photos you’ve taken and screenshots you’ve captured. That latter part opens up some pretty interesting possibilities. Say, for instance, you’ve just gotten an email with a tracking number in it, but the tracking number is some funky type of text that annoyingly can’t be copied. (This seems to happen to me way too often.) Or maybe you’re looking at a web page or presentation where the text for some reason isn’t selectable. Well, grab a screenshot — by pressing your phone’s power and volume-down buttons together — then make your way over to the Google Lens app. Tap the image icon in Lens’s lower-left corner, look for your screenshot in the gallery that appears, then tap it. And from there, you can simply touch your finger anywhere on the screen to select any text you want. (The same capability is also now present in the newer Android Circle to Search system, by the by, though that feature is still much more limited in its availability.) You can then copy the text, send it to a computer, or perform any of Lens’s other boundary-defying tricks. Speaking of which… Google Lens trick #7: Search for any text, anywhere After you’ve selected any manner of text from within the Google Lens app, look for the “Search” option within the pop-up panel that appears atop it. It’s all too easy to overlook, but alongside the other options we’ve gone over sits the simple and supremely useful “Search.” Keep that option in mind as a super-easy way to get info on text from any physical document or captured image without having to manually peck in the words on your own. (Sometimes, Lens will even put related results right within a panel beneath your image, without any additional searching required.) And on a related note… Google Lens trick #8: Search for similar visuals We already know that Lens can search for the text from an image. But the app is also capable of searching the web for other images — images that match the actual objects within whatever photo or screenshot you’re viewing. It’s a fantastic way to find visually similar images or even identify something like a specific phone model or product seen within a photo. To pull off this slice of Googley sorcery, open up an image of anything within Lens — or just point your phone at an object in the real world — then swipe up on the panel that appears beneath it and look for the “Visual matches” section. Searching for similar visuals is a smart way to get extra context about anything around you.JR Raphael, Foundry Google Lens trick #9: Save someone’s contact info If you find yourself holding a business card and thinking, “Well, blimey, I sure as heckfire don’t want to type all of this into my contacts app,” first, congratulate yourself on the excellent use of blimey — and then sit your beautiful person-shell back and let Lens handle the heavy lifting for you. Open Lens, point your phone’s camera at the card, and tap on the person’s phone number or email address. The Google Lens app should recognize the nature of the info and prompt you to add a contact. One more tap, and the deed is done. Google Lens trick #10: Email, call, text, or navigate Got an address or number you need to get onto your phone for a specific sort of action? It could be on a business card, on a letter, or even on the front of a random business’s door. Whatever the case, just open the Google Lens app, point your phone at it, and tap the text. (Or, option B: Snap a photo of the info in question and then pull it up in the Lens app later.) Once Lens sees it, it’ll offer to do whatever’s most appropriate for the sort of info involved. Then, with a single tap, you’ll have the address ready to roll in a new email draft, the number ready to call or text in your dialer or messaging app, or the website pulled up and ready for your viewing in your browser — no time-wasting typing required. Google Lens trick #11: Translate text from the real world If you ever find yourself staring at a sign in another language and wondering what in the world it says, remember that the Google Lens app has a built-in translation feature. To find it, open the app, aim your phone at the text, and tap the word “Translate” along the bottom edge of the screen. Before you know it, Lens will replace the words on your screen with their English equivalents (or with a translation in whatever language you select, if English isn’t your tasse de thé) — practically in real time. It’s almost spooky how fast and effective it is. srcset="https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?quality=50&strip=all 800w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=289%2C300&quality=50&strip=all 289w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=768%2C798&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=671%2C697&quality=50&strip=all 671w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=162%2C168&quality=50&strip=all 162w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=81%2C84&quality=50&strip=all 81w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=462%2C480&quality=50&strip=all 462w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=347%2C360&quality=50&strip=all 347w, https://b2b-contenthub.com/wp-content/uploads/2025/04/google-lens-android-translate.webp?resize=241%2C250&quality=50&strip=all 241w" width="800" height="831" sizes="(max-width: 800px) 100vw, 800px">Translation on demand, courtesy of Google Lens.JR Raphael, Foundry Pas mal, eh? Google Lens trick #12: Calculate quickly The next time you’ve got a numerical challenge in front of your weary peepers, give your musty ol’ brain a break and let Lens do some good old-fashioned solvin’ for ya. Just open up Lens and point your phone at the equation in question — whether it’s on a whiteboard, a physical piece of paper, or even a screen in front of you. Scroll over along the line at the bottom of the Lens viewfinder screen until you see the word “Homework” (and don’t worry: Despite what that label implies, you don’t have to be an annoyingly youthful and bushytailed student to use it). Tap that, then tap the big Lens search icon. And with everything from basic equations to advanced math, chemistry, physics, and biology, Lens will eagerly do your calculation for you and spit back an answer in the blink of an eye. I won’t tell if you don’t. Google Lens trick #13: Scan your skin Here’s a weird one: If you ever have some mysterious marking on your mammal skin and find yourself fretting over whether it’s a freckle or something more nefarious, take matters into your own hands and let Lens play the role of dermatologist for you. Just fire ‘er up on whatever Android phone you’re holding and point the viewfinder at your >wretch-inducing wart perfectly natural epidermal abnormality, then tap that Lens search button. Before you can spit out the words “Holy moley,” Lens will give you a best guess at what you’ve got goin’ on. Its results aren’t scientific, of course, but they are based on matching your marking to an endless array of examples Lens seeks out on the web — so they’re a fine way to put your mind at ease while you wait for an actual doctor to examine your suspiciously spotted outer layer. Google Lens trick #14: Crack the codes ‘Twas a time when Android code-reading apps were all the rage — and plenty of folks still have ’em hangin’ around today. So long as the Google Lens app is on your phone, though, guess what? You don’t need anything more. Just open up Lens, aim your camera at any barcode or QR code, and poof: Lens will offer to show you whatever that code contains faster than you ask “What does QR stand for, anyway?” Who needs a QR code reader when Lens is ready and waiting?JR Raphael, Foundry Being a mobile-tech magician has never been so satisfying. Get six full days of advanced Android knowledge with my free Android Shortcut Supercourse. You’ll learn tons of time-saving tricks for your phone!

The Google antitrust trial is “a nonsensical move by the US government that is ridiculously short-sighted, because if compromising privacy for ad revenue is bad, compromising privacy to train AI models is dangerous,” an industry analyst said Thursday. Paddy Harrington, senior analyst at Forrester, who specializes in security and risk, added that efforts by the US Department of Justice (DOJ) to derail the company is a “move that may have been done in the name of anti-trust and competition, but it does not appear as if anyone thought through the implications of something else.” That ‘something else’ raised its head this week during the remedies portion of the trial that began in August 2024, in which Judge Amit Mehta ruled against Google. Nick Turley, the product lead of ChatGPT at OpenAI, said his company would be interested in acquiring Google’s Chrome browser if the company is forced to sell it. Could disrupt the browser market Harrington countered, “this is not as simple as selling off a product; it’s a complete platform. And it’s moving from Google, where data collection is about selling ads, to OpenAI, where data collection is about training AI to then sell to a ridiculously wide variety of purposes. A ‘devil you know versus the devil you don’t know’ sort of deal.” Such a move, said Harrington, has the potential to completely disrupt the browser market as a whole, not just for Google. “It’s understandable why OpenAI would want it for training the AI models, but if they purchase Chrome, what happens to Chromium? While it’s the open-source project of Chrome and ChromeOS, does Google keep the project under their development arm, or does it go with the browser?” He pointed out that the majority of non-Chrome browsers use Chromium as their engine, so “if it stays under Google, by and large the browser market should probably remain the same. If it goes along with the sale to OpenAI, that could cause a serious disruption, as the privacy focused developers may want to distance themselves from a company that’s making money feeding their AI models.” It’s all about the data OpenAI’s interest in Chrome, said Brian Jackson, principal research director at Info-Tech Research Group, “is all about the data. OpenAI acquiring Chrome would be like Exxon acquiring an untapped oil field with the drilling already done. Just turn on the taps, and the good stuff starts flowing. Beyond that, OpenAI also has about 3.5 billion users, which it could reach instantly with its AI services.” It is, he said, possible to imagine an OpenAI version of Chrome that would start with ChatGPT search by default instead of Google.com, and that would offer agentic integration to complete tasks such as shopping, booking appointments, or unsubscribing from unwanted emails. “Such a coup would instantly transform OpenAI’s outlook from a provider of LLMs that sit behind the scenes of major tech brands to one of most prominent direct owners of customer relationships in the world,” said Jackson. “ChatGPT is already propelling it into that territory, but Chrome is still many magnitudes more well used.” As for the impact all of this might have on enterprise users, Harrington said that “it would make sense for Chrome Enterprise (CEB) to go along with the ‘consumer’ Chrome business, and while there are two editions — one free, one paid — the paid version really just opens up the number of functions you can manage inside the browser. The functions are there already, it’s just a matter of how they’re turned on and off.” All of this, he said, “is tied to Chrome itself, and Chrome Enterprise is just a rebundled version of it into an MSI (for enterprise distribution and an enterprise approach to configuration) with ADMX templates.” Anshel Sag, principal analyst at Moor Insights & Strategy, agreed, saying he could not imagine Chrome Enterprise being sold separately from the consumer version, “since they are fundamentally the same browser with different features and security.” A question of trust Asked about the level of trust CIOs and CISOs have in Google compared to OpenAI, he said, “there is more trust in Google, simply because the company has shown that it’s serious about security and privacy. That said, OpenAI doesn’t really have that kind of a track record, and while it does work with enterprises, it is still unclear how it might handle something like a browser, and all of the responsibilities that come with maintaining that level of security.” One of the biggest costs that would further impact OpenAI’s profitability is R&D for Chrome, and ensuring it remains secure, Sag added. Jackson said there is no doubt Chrome Enterprise would be impacted. The service, he said, is “the browser with some other enterprise management capabilities wrapped around it. It’s possible OpenAI would reach an agreement to still allow Google to own and sell its enterprise contracts there, but the product at the core of it would still be controlled by OpenAI.” Then, he said, “there’s the question of what OpenAI does with the data, and how the change in ownership would impact enterprise trust. When it comes to cybersecurity in the enterprise, the modern best practice approach is a zero-trust approach. Zero-trust principles say to treat all connections as potentially hostile, limiting access to systems and data based on only what is absolutely needed by an authenticated user.” The ’devil we know’ For browsers, he said, “ there are a couple of ways that enterprises try to make that work. Some large enterprise clients will use Chrome Enterprise, which offers browser policy management. This is where they can limit what data the browser can access and what it can do with that data.” Other enterprises, said Jackson, would let users download pre-approved consumer browsers, likely including Chrome, and have IT manage them with other centralized administration tools such as Windows Active Directory or a Unified Endpoint Management solution. He said, “unless OpenAI tries anything too aggressive with data collection, we’ll see enterprises take the same approach to web browsers in the future. Ultimately, Google is also a large data-hungry tech giant that is trying to train its own AI models, and we’d have every reason to be as concerned about its motivations to harvest user data as OpenAI. But it’s the devil that we know.”

As previously reported, several European countries are concerned that the Chinese-made DeepSeek chatbot is collecting sensitive data about users without their consent. Now, Reuters reports that South Korea is also investigating the issue, after previously blocking access to the app in the country. According to South Korea’s data protection authority, DeepSeek passed on user data to a number of different Chinese companies before the app was blocked — and both user data and the content of users’ prompts were sent to Volcano Engine Technology, a company based in Beijing. DeepSeek has defended itself by saying that the data was passed on to improve the user experience, but this explanation is not accepted by South Korea’s data protection authority. The Chinese company also says it stopped forwarding the content starting April 10. DeepSeek is now being asked to immediately delete the data collected from South Korean users.

During the pandemic, remote working was common. But many large tech companies have changed their tune in recent years, insisting that their employees to return to the office for at least part of the workweek — even when they lack adequate office space. The latest news is that Google is now threatening to lay off employees in certain departments who are not in the office at least three days each week. For example, remote workers in the Technical Services unit were told they must choose between a hybrid work schedule and a voluntary exit package, according to CNBC. Employees in the unit who live more than 50 miles from the office will be offered financial compensation to move closer to work. This follows an internal memo Google co-founder Sergey Brin sent in February, in which he urged employees working on the company’s Gemini AI model to be in the office every weekday and work 60 hours per week. “Face-to-face collaboration is an important part of how we come up with new inventions and solve complex problems,” Google spokesperson Courtenay Mencini told CNBC. Previously, tech giants Meta, IBM, Apple, Microsoft, Amazon, and Tesla, among many others, have chosen to limit the right to telework, despite numerous studies showing that RTO mandates hurt worker productivity, satisfaction, and retention.

Sadly, there are many reasons why data stored on disk drives in Windows can — and sometimes does — go missing. As someone who’s used Windows 11 since the day after the first Insider Preview appeared on June 28, 2021, I’ve been through a dozen different data recovery incidents over the past 3-plus years. Some of these have resulted from failing storage media, others from issues with cables or connectors between the media and the PC or its motherboard, and one or two have been self-inflicted arising from mistakes when using disk partitioning or repair utilities. What can I say? I like to experiment, and break things so I can fix them. In each of these incidents I’ve been able to claw back my missing data, and in most cases you can too. In this story I’ll go over several techniques you can use to try to recover lost files. I’ll use Windows 11 in the examples throughout, but these tips apply to Windows 10 as well. In this article: Three main reasons for data loss When recovery is needed, backup’s the thing Step 1 to data recovery: Recycle Bin Step 2: Windows Backup / File History Step 3: chkdsk Step 4: Windows File Recovery Step 5: Partition/data recovery If file recovery succeeds, follow up with a disk health check Step 6: Call in the professionals As with my tutorials for repairing Microsoft Office and repairing Windows itself, the idea is to perform these steps in order. If the first step doesn’t work, move on to the next, and so forth. With luck, you’ll recover your missing data early in the sequence. But first, let’s take a look at the main culprits when it comes to Windows data going MIA. Three main reasons for data loss In most cases, three primary causes stand behind situations that call for data recovery. These are: 1. File or file structure damage or corruption. Basically, files are just long sequences of bits (1s and 0s) chained together through a master file table (or its equivalent) on the storage media. For various reasons, those patterns can become damaged or disrupted. Consequently, they may go missing or become unreadable. In this case, your repair efforts attempt to restore those bits to their original form and order. 2. Partition or disk structure changes or damage. In Windows, storage devices use specific disk layouts to create logical volumes (which usually have names and associated drive letters) wherein file systems reside. Each volume (and certain other disk structures related to boot-up and recovery on Windows system/boot disks) has at least one associated on-disk partition. The drive maintains one or more copies of a partition table to keep track of disk extents, called sectors, where partitions (volumes and other special disk regions) reside. When errors or unwanted changes affect a partition table, it must be repaired before the files and folders in any of the volumes it describes may be accessed. In this case, the repair process attempts to restore the partition table’s original structure and contents. 3. Storage hardware problems. For spinning disks, this might involve damage to a disk platter’s read/write heads or one of its recording surfaces, and usually manifests as loss of drive sectors. In severe cases, a storage device may fail completely so that its contents become completely unreadable. For SSDs, something similar can happen when the drive’s controller suffers from hardware errors, or when the flash memory locations under its control become damaged or unreadable. For both kinds of drives, failure is often a progressive cascade. That is, it starts slowly, with the failure of one or two sectors or memory locations, and gets progressively worse over time. Eventually the device fails completely. At that point, end-user recovery is impossible, and one must turn to a data recovery company for expensive professional recovery services. That’s why it’s best to act upon the first sign(s) of disk trouble while there’s still time to head problems off at the pass. In this case, repair means attempting to capture valid data from good sectors or locations before they have a chance to go bad. When recovery is needed, backup’s the thing Long before you encounter data loss issues, it’s worth employing one or both of Windows’ built-in file backup tools. File History takes snapshots of files stored in certain main folders in Windows 10 and 11. Windows Backup saves your files, apps, settings, and credentials to OneDrive in the cloud. These set ‘em and forget ‘em tools require little or no effort to use once set up; see my tutorials on File History and Windows Backup for details. In my opinion, though, nothing beats access to a fresh, current system backup of a drive from which items may be missing or lost. For example, I keep project work on two separate SATA SSDs (F: and X:) and back them up every morning, along with the all-important C: boot/system drive where Windows itself lives. Thus, I’m reasonably sure that no matter what happens to one or more of those drives, I won’t lose more than a day’s work or data, worst case. Before attempting any kind of file or data recovery on Windows 11, the first thing you should do (let’s call it Step 0) is back up the problem drive on which missing or inaccessible files or data once resided. This means you can always get back to where you started, should your own data recovery efforts cause further problems. You will then be able to turn the media over to a data recovery company that has tools and capabilities that ordinary users like you, dear reader, and I do not have at our disposal. That’s the worst-case scenario, though. There are several steps you can take first to try to recover your lost files yourself. Let’s walk through those first, shall we? Step 1 to data recovery: Recycle Bin A typical situation that calls for data recovery is one where a user or some program has deleted a file from a Windows volume, intentionally or perhaps accidentally. Afterward, the user wishes to regain access to that file. It seems obvious, but checking Recycle Bin to see if the missing file is present can often save a lot of unnecessary hassle when recovering lost items. Double-click the Recycle Bin icon on the desktop, then browse or use the search box to see if the missing file is there. If it turns up, a right-click on the item, then selecting Restore is all that’s needed to put the file back where it came from (see Figure 1). Figure 1: Right-click any item in Recycle Bin, then click Restore. That’s all it takes to recover files from the Recycle Bin. Ed Tittel / Foundry If the file you seek is not available in the Recycle Bin, other actions may help. Step 2: File History / Windows Backup This step comes with some ifs attached. The first if is “Is File History (or Windows Backup) turned on?” The second if is “Does File History (or Windows Backup) include this file?” If so, the third if is “Are the damaged or missing files available through File History (or Windows Backup)?” To briefly explain: both File History and Windows Backup are turned off by default and must be enabled. Even then, these utilities only back up files stored in specific folders — namely, Documents, Downloads, Pictures, Videos, and Music (a.k.a. “Library folders”) from the logged-in user’s file hierarchy. (See my File History and Windows Backup tutorials for details on setting them up, including a way to include additional folders in File History backups.) If you have both tools enabled, consider what kind of restore operation you need. Windows Backup offers basically an “all-or-nothing” restore capability. If you’re simply looking to restore one or a small number of files and/or folders, it’s not the right tool for the job. Why? Because Restore in Windows Backup works through the Windows Installer — that is, when you start the Windows install process and log in with your Microsoft Account (MSA), it will offer to restore the PCs for which it has backups associated with that account through OneDrive. This is generally best reserved for situations where a full system restore or rebuild is called for. It’s not a good fit for restoring some smaller number of files or folders. If you take this route, please understand you’ll have to wait through a clean Windows 10 or 11 install, after which Windows Backup will start the process of bringing your apps, files and folders, settings and preferences back (but there will still be more work for you, as explained in my Windows Backup tutorial — see the section “WB comes with a catch…”). For a one-off file or just a handful of items, use File History instead.  Click Start > Settings, then type file history into the Settings search box. From the options that appear, select Restore your files with File History. You’ll see a File Explorer-like view of your backed-up folders. Click any folder to drill down to the item you want. If you navigate into the Documents folder, for example, you’ll see its contents show up. By highlighting one or more files, you select them for a “copy back” operation. Then, when you click the green arrow at bottom center (see Figure 2), it will restore the files you’ve selected to their original location. Figure 2: Select an item and click the green “back-circle” arrow to restore it. Ed Tittel / Foundry For in-depth details about File History, including its limitations, setup minutae, and the best way to configure it, see “How to use File History in Windows 10 and 11.” If a file or folder is not available for restoration through File History or Windows Backup, more robust recovery is needed. That means it’s time to move on to the third step. Step 3: chkdsk If the affected drive remains readable and visible in Windows — that is, the drive letter still appears in File Explorer and some (or most) of its files and folders remain visible therein — the next step to recovery is to use the command line check disk utility, a.k.a. chkdsk.exe. You’ll need to run it at an administrative command prompt or in an administrative PowerShell session. (Type either cmd or powershell in the Windows search box, right-click Command Prompt or Windows PowerShell in the list of results, select Run as administrator, and enter the password if prompted.) You can use either of two key parameters to effect disk repairs, as outlined on the MS Docs Windows Commands reference for chkdsk): /f: Fixes errors on the disk. The disk must be locked. If chkdsk cannot lock the drive, a message appears that asks you if you want to check the drive the next time you start the computer. /r: Locates bad sectors and recovers readable information. The disk must be locked. /r includes the functionality of /f, with the additional analysis of physical disk errors. The only way to repair the Windows boot/system disk (usually drive letter C:) is to elect the option to repair after the next restart, followed by a restart of the affected PC. For minor disk corruption or damage, this step is often as far as you’ll need to go. If it works, count yourself lucky — and make another backup right away! Step 4: Windows File Recovery Microsoft offers a good, basic file recovery utility called Windows File Recovery (winfr.exe) for Windows 10 version 2004 and up and all versions of Windows 11. It’s available in the Microsoft Store. This utility launches in an administrative command prompt session and runs at the command line. I tried the command to recover download files, as shown in Figure 3. Figure 3: winfr.exe is added to the path, and online examples are entirely worth trying out, as shown for downloads here. Ed Tittel / Foundry There is a learning curve involved in using this tool, but the user manual is available online and very much worth reading and exploring for those willing to tackle file and folder recovery at the command line. If you know what’s missing and where it resided, you can use it to recover files quickly and easily using specific filters — the more specific, the better, as I learned when recovering my old download files. The winfr tool also supports what’s called “extensive mode,” which offers more comprehensive and far-reaching repair capabilities. Regular mode is turned on by default; you must use the /extensive parameter to access extensive mode’s capabilities. It’s good at handling non-NTFS file sytems (e.g., FAT and exFAT), and can find more files deleted longer ago. Extensive mode can also attempt repairs after formatting a disk — which takes us into territory for the next topic, partition repairs — and it is better able to cope with disk corruption than regular mode. Those who prefer to undelete files using a graphical interface instead of working at the command line should try the free but good Piriform tool Recuva. Step 5: Partition/data recovery There are plenty of good third-party data recovery tools available for Windows 11. That said, those that can recover entire drives — especially big ones — or that offer partition and formatting recovery usually cost money. For example, upgrading from the free version of Recuva to the Professional version means gaining more advanced file recovery, including partition recovery, recovery from damaged or reformatted disks, and more extensive recovery for long-deleted files and folders. It costs US$25 for a one-year subscription. Similarly, MiniTool Partition Wizard, which offers excellent partition handling and modest partition recovery in its free version, but adds extensive file, folder, and partition recovery capabilities to its Pro version (US$59 yearly). Ditto for Stellar Data Recovery, which comes in a variety of versions that range from Standard (US$60 yearly) to Toolkit (US$299 yearly/$599 lifetime). Only those who need to recover BitLocker-encrypted or multi-OS drives need those higher-end versions. In my testing, the Toolkit version found 2.4TB of recoverable files on a 1.84TB drive, which strikes me as both impressive and entirely too comprehensive. You can see that in the lower left corner of Figure 4. Figure 4: Remarkably, Stellar Data Recovery found 2.4TB of recoverable files on a 1.86TB drive. Ed Tittel / Foundry Stellar uses a simple File Explorer-like UI through which you can navigate to select files or folders to restore from its collection of recoverable items. It’s intuitive and easy to use, and quite fast, considering what it does (took less than 10 minutes to offer a huge list of recoverable files and folders, as shown). Once partition or file recovery is effected, you can then tackle file and folder recovery from newly restored partitions (if needed), or from the list of recoverable items your tool offers up. In cases where accidental reformatting or partitioning mistakes have been self-inflicted — I’ve done this to myself on at least two occasions when I targeted the wrong drive — successfully restoring the old partitioning scheme usually brings all the old files and folders back, too. If file recovery succeeds, follow up with a disk health check Something had to cause whatever issues led you down the recovery road. If you’ve managed to recover data through chkdsk, Windows File Recovery, or another data recovery tool, it’s a good idea to give the problem disk a thorough checkup after that’s done. Hard drive and SSD makers often offer such tools as downloads for the drives they sell, and there are plenty of free and for-a-fee tools available for disk health checks. I’m partial to the open-source CrystalDiskInfo because it provides a good overview of a disk’s SMART data (self-monitoring, analysis and reporting technology, which includes error counts that often signal impending disk failures). Other tools such as HD Tune Pro ($35, occasionally available for free, plus free trial) and Hard Disk Sentinel (free trial available, $20 for the standard version, $30 for the pro version) offer more detailed disk diagnosis and surface/media scans to provide better information on disk health and condition. The former took 30 minutes to scan my 3-year-old Western Digital SN810 SSD (nominal 2TB; actual 1.86TB) before giving 80% of it a clean bill of health despite heavy usage, as shown in Figure 5. (The free trial won’t scan the full disk as an inducement to buy the Pro version.) Figure 5: The HD Tune trial version error scan covers only 80% of any disk to persuade users to buy a Pro copy. Ed Tittel / Foundry If a disk shows questionable health or more than 10% of its sectors are damaged, it’s time to replace it. Make a backup while you still can, order a new device, and you’ll be able to restore the backup when the replacement arrives. Step 6: Call in the professionals If steps 1 through 5 fail, you’ll need to seek outside help. Here’s Software Testing Help’s list of the top 10+ data recovery services for those who must travel this road. Hopefully, you’ll never need to send a drive to a professional data recovery service. If you do, be prepared to pay handsomely for their work. Charges of $300 and up are typical. (Bigger drives cost more, as you might expect.) And it can take weeks or months to get through their service queues. Some companies will let you send them a replacement drive on which to restore what they find. Others will insist on selling you such a drive, sometimes at above-market prices. But sometimes, there is no alternative, and you’ll pay what you must to get precious or irreplaceable files and data back. Be sure to work out as much of the cost picture as you can before sending a drive off for recovery. You don’t want the bill, however big it turns out to be, to come as a surprise. In my 30-plus years of working with personal computers, I’ve never had to send a drive off for professional recovery. Even so, it’s good to know such services exist should one ever need them. Personally, I learned the value of backups in 1989, when a 300MB SCSI drive attached to a Macintosh cratered. I lost a book manuscript as a result. I’ve never been caught unprotected again, nor should you be. Happy computing! This article was originally published in June 2021 and updated in April 2025.

If it looks like a trade war, swims like a trade war, and quacks like a trade war, then it’s probably a trade war that has now broken out — this time between the US and the EU as the White House condemns Europe’s punitive fines against Apple and Meta, fines the companies intend to appeal. Europe hit Apple and Meta with fines of €500m and €200m, respectively, yesterday, punishing both companies for noncompliance with Europe’s Digital Markets Act, a piece of legislation that pretends to be about opening up markets but seems custom-designed to impact the US tech giants. The White House has called these fines a “novel form of economic extortion” and has warned Europe that the US will not tolerate the magnitude of these fines. Custom-fitted penalties The steep fines surprised most commentators, as whispers coming out of the bloc had hinted that the EU would impose minimal fines against both tech companies in order to avoid reprisals from the US administration. This doesn’t seem to be what happened, unless we assume that €700m is now seen as small change by Europe’s leaders, who appear to have given themselves the right to charge US companies even more.  The fines come as Europe and the US attempt to forge new trading agreements in response to pressure from the US administration and its tariff threats. While the impact of tariffs will hurt US consumers most, repercussions will also be felt by manufacturers and trading partners who have fed US demand until now. The effect will also soon be felt on a wider basis as a trade barrier-induced slump hits shipping and distribution globally in the coming weeks, reflecting the slowdown in demand around the imposition of those tariffs.  Playing at leadership Despite the looming risk of consequences for their own economies, both in terms of manufacturing demand and the impact on their own manufacturing businesses of an onset of low-cost consumer goods originally destined for the US market, Europe’s leaders seem to want to cosplay at playing hardball. European Commission spokesperson Thomas Regnier recently said the EU “will enforce our tech legislation without any doubt, and this has nothing to do with the trade negotiations currently ongoing with the US.” That may be how Europe’s leaders see it, but their self-perception means little to a White House that sees these fines as extortionate reprisals against some of America’s biggest and most successful firms. The administration is far more likely to cleave to the opinion of Meta and Apple: “The European Commission is attempting to handicap successful American businesses while allowing Chinese and European companies to operate under different standards,” Meta said. “Today’s announcements are yet another example of the European Commission unfairly targeting Apple in a series of decisions that are bad for the privacy and security of our users, bad for products, and force us to give away our technology for free,” Apple said. Who designed the rules, and for what purpose? While Regnier insists the rules are being applied fairly and would be applied against any firm, no matter where they are from, critics argue that the DMA seems to have been expressly drafted to constrain the power of US firms. The US is not blind to this conjecture. “Extraterritorial regulations that specifically target and undermine American companies, stifle innovation, and enable censorship will be recognized as barriers to trade and a direct threat to free civil society,” said a White House spokesperson. The bellicose response emerging from within the US administration suggests it is quite willing to issue its own set of reprisals against Europe’s attempts to fine the tech firms — but that response may not be immediate, pending the result of any legal appeals to those decisions on the part of Apple and Meta. The scenario doesn’t equate to the best mood music. At a time when the administration is practicing a very blunt approach to making deals, European leaders seem to want to hide the true nature of their own equally self-serving responses behind unconvincing veneers of respectability (such as the DMA). Will the future be better tomorrow? In this kind of context, the idea that relationships may become worse before they get better isn’t just a problem waiting to happen. It appears to be a problem that’s already here. Like a squawking duck, this particular sequence of events certainly seems to be forming up to become exactly what it sounds like as the US administration puts its own perception of national interest first, unleashing a challenging set of circumstances for businesses worldwide, including those of Apple and Facebook, the US business entities it probably sees itself as trying to protect.  Ironically, the consequences of these combined sequences of events probably won’t unleash a great deal of benefit for anyone — except, possibly, for some wealthy individuals who want to sell apps via their own App Store. US President George W. Bush once famously said, “The future will be better tomorrow.” Right now, in the absence of positive dialogue, that’s not what I see coming down the pipe. You can follow me on social media! Join me on BlueSky,  LinkedIn, Mastodon, and MeWe. 

A growing chorus of European technology executives is calling for the continent to assert control over its digital future, with Danish IT services giant Netcompany leading the latest push for technological self-reliance. In an open letter published on Wednesday — coinciding with the symbolic illumination of the Statue of Liberty replica in Paris a day before — Netcompany CEO André Rogaczewski explicitly challenged Europe’s dependence on foreign technology platforms and urged the region to “bring our data home.” “From social media to cloud infrastructure, from applications to algorithms, we are dependent on technologies developed elsewhere, by actors who may not share our values,” warned Rogaczewski, whose firm employs over 8,000 technology consultants across Europe. The high-profile campaign comes amid escalating tensions in global technology markets, with the European Commission recently intensifying scrutiny of US cloud providers’ market practices and ahead of upcoming EU-US discussions on trans-Atlantic data governance frameworks. “We are calling for European solutions — built by European companies, run on European data, and accountable to European citizens,” Rogaczewski stated, directly challenging the market dominance of American tech giants including Microsoft, Google, and Amazon Web Services, which collectively control a significant majority of Europe’s cloud infrastructure market according to industry reports. Europe’s strategic pivot in digital policy The remarks come amid a concerted push by European governments and institutions to localize control over key digital systems. Recent EU policies — the Digital Services Act, the Digital Markets Act, and the AI Act — are part of an evolving legal framework to strengthen regional oversight of platforms, algorithms, and cloud-based services. A month ago, leading European companies and lobbying groups — including Airbus, Element, and Nextcloud — under the umbrella of “EuroStack Initiative” signed an open letter urging the creation of an EU sovereign infrastructure fund to boost public investment in innovative technologies and build strategic autonomy in key sectors. “Building strategic autonomy in key sectors is now a recognised urgent imperative across Europe. As part of this common effort, Europe needs to recover the initiative, and become more technologically independent across all layers of its critical digital infrastructure,” the EuroStack letter read. These initiatives follow a global trend where technology is no longer seen purely through the lens of innovation or efficiency but as a strategic national asset. The US has tightened its grip on semiconductor exports to China. China, in turn, is accelerating its own domestic tech stack and enforcing data localization. In this shifting context, Europe’s historical reliance on the US and Chinese digital infrastructure has become a liability. Building a European tech ecosystem Netcompany, a publicly listed IT services provider with operations across Europe, is among a growing number of regional firms advocating for digital sovereignty. Their CEO’s comments underline the urgency to reduce reliance on US-based cloud giants and software vendors. Instead, the letter encourages a continental effort to cultivate indigenous technologies that align with European legal standards and ethical norms. “Technology lies at the heart of our wealth creation,” Rogaczewski said. “It drives our competitiveness and sits at the very center of how we communicate, learn, and develop as societies.” This vision extends beyond public discourse into concrete initiatives. GAIA-X, a European cloud infrastructure initiative, exemplifies this push toward a sovereign tech ecosystem, alongside other strategic programs including SiPearl and the EU’s AI Continent Action Plan that target capabilities in cloud infrastructure, semiconductors, and AI. US tech giants have not been idle in response to these sovereignty concerns. Amazon Web Services, for instance, has committed to a €7.8 billion ($8.9 billion) investment in an “AWS European Sovereign Cloud” and maintains that its approach has been “sovereign-by-design” from the beginning, with customers having “complete control over where they locate their data” within European regions and verifiable control over who can access it. “While complete technological independence is a complex and long-term goal, Europe is clearly building momentum toward digital and AI sovereignty,” said Shreeya Deshpande, senior analyst at Everest Group, highlighting how the coordinated nature of such efforts reflects growing momentum across Europe. While challenges remain, particularly in scaling and integrating across fragmented markets, the political will and regulatory backing for European tech nationalism is growing. Sovereignty without isolation Rogaczewski’s appeal reflects a growing consensus among European stakeholders that sovereignty does not mean isolation. Rather, it signals a recalibration of Europe’s role in the global digital order. Europe is seeking to maintain open markets and innovation, while ensuring that core digital infrastructure and sensitive data remain under regional control. “Emerging mechanisms, such as data embassies and sovereign cloud frameworks, offer a practical middle path — enabling countries to maintain legal and operational control over data and AI systems while remaining interoperable with global platforms,” Deshpande added. The message resonates with policymakers who see technology not just as a tool of commerce but as a pillar of democratic governance. “This places our security, sovereignty, and democracy at risk,” Rogaczewski warned, referring to Europe’s current dependency on foreign platforms. The lighting of the Statue of Liberty — once gifted by France to the US as a symbol of shared democratic values — served as a potent backdrop to Rogaczewski’s message. He framed his letter as both a reminder of historical ties and a warning that those values are now “under heavy pressure.” “Our modern societies are based on the very same principles of freedom and democracy,” he wrote. “We must stand united in our commitment to these values and fight for them each and every day.”

Microsoft is strengthening its Copilot generative AI software as the centerpiece of the workspace by adding key features to help users be more productive. The company’s next wave of Copilot updates includes “digital labor” tools to automate work and merge document intelligence, with the aim of helping enterprise users to work smarter, not harder. For example, a new Copilot Search feature neatly summarizes information by going through wads of unstructured information stored in an organization’s apps and systems. Copilot can provide context and present information after searching for data in common Microsoft productivity tools and third-party apps that include Google Drive, Slack, ServiceNow, and software from Atlassian. The new AI features, which include other summarization, desktop personalization, and administrative tools, are possible because of “advanced models, adaptive memory, and reasoning agents,” Microsoft said in a blog post. Other new Copilot features include “memory and personalization” capabilities — something Microsoft touted at its recent 50th anniversary event — which can personalize the desktop experience by retaining and understanding user preferences over time. “Memory and personalization are private just to you, and if you’re ever working with sensitive information, you can control what Copilot remembers,” Microsoft wrote in its blog. The “Copilot Notebooks” feature allows users to create a mini-AI model based on their notes and documents. Users can create a “notebook” and load it with relevant documents based on the task at hand. The notebook will then present contextual answers by scanning only those documents. The notebooks can be updated with new documents, which the AI model indexes into the corpus of data. Copilot Notebooks is similar to Google’s NotebookLM app, which is included in the competitor’s Workspace productivity suite. Google also announced the addition of new genAI features to Workspace at this month’s Cloud Next conference. The new capabilities will begin rolling out to Microsoft 365 Copilot customers in late May, the company said. Microsoft’s paid enterprise genAI offering is available as an add-on to commercial M365 subscriptions for an additional $30 per user per month. In Microsoft’s new Agent Store, customers can find pre-built AI agents from Microsoft and partners including Jira, Monday.com, and Miro. Microsoft’s previously announced Researcher and Analyst reasoning agents are rolling out to customers in the new Agent Store starting today via the company’s Frontier program. Companies can also offer their own custom agents in the store. Microsoft also announced the Copilot Control System for system administrators to manage the creation and deployment of AI agents within enterprise systems — including the ability to enable, disable, or block agents for specific users or groups. Microsoft is at the very beginning of the journey toward agentic AI, said J.P. Gownder, vice president and principal analyst at Forrester Research. “Microsoft Copilot Agents are good for tapping into specific data sources — like a SharePoint site — to ensure that you are querying the right data source,” Gownder said. Forrester views some Copilot features that handle simple tasks and tap into specific sources as being more “agent-ish” as opposed to true agentic AI. “Agent-ish solutions can solve point problems, but they aren’t the ‘digital coworkers’ that vendors are talking about, at least not yet,” Gownder said. As Copilot and other AI systems further integrate with enterprise tools, system administrators will need to put in guardrails so unauthorized users aren’t able to access sensitive information from AI features, said Jack Gold, principal analyst at J. Gold Associates. “Do you let Jack, who is a low-level employee, ask anything he wants? Knowing roles and security and data access clearance of various users… is going to be very important for general-purpose AI systems,” Gold said. Users will also need to use common sense, as AI tends to wander if not given specific context or criteria in queries. Depending on the role and data complexity, the user training process “will probably be a trial-and-error issue,” Gold said. “The biggest challenge will be: can you trust the results that the agents just provided you?” Gold added.

Intel has announced a workforce reduction of over 20%, impacting more than 21,000 employees, as part of a restructuring led by new CEO Lip-Bu Tan, per Bloomberg. With a $19 billion loss in 2024 and revenue falling to $53.1 billion, the chipmaker aims to streamline management, cut $10 billion in costs, and refocus on AI chips and foundry services, competing with Nvidia, AMD, and TSMC, the report said. Tan’s restructuring, the biggest workforce reduction in the chipmaker’s history, follows 2024’s layoff that impacted 15,000 employees, bringing the headcount down to 108,900 from 124,800. The current cuts, targeting middle management, reflect Tan’s push to restore an engineering-led culture and reposition Intel amid mounting pressure. Once a semiconductor icon, Intel is undergoing a radical shift to meet AI-driven demands and modern IT infrastructure needs. In his efforts to streamline Intel, Tan is shedding non-core assets, starting with the sale of a 51% stake in Altera to Silver Lake. He said the move reflects Intel’s commitment to sharpening its focus and lowering its expense structure. Tan also highlighted the need to rebuild engineering talent, strengthen the balance sheet, and better align manufacturing with customer demands. Intel’s recent promotion of Sachin Katti to CTO and AI chief reflects urgency to close the AI gap, though analysts view these executive changes as part of a “monumental uphill battle.”  These moves mirror a broader trend in tech, with more than 22,000 jobs cut in 2025 alone, continuing the 2024 wave of layoffs. The shift towards cost-cutting and AI automation signals rapid changes in the industry. Remaking a chip giant The global semiconductor market is undergoing a dramatic shift, driven largely by the AI boom. As per Gartner, it hit $655.9 billion in 2024 — up 21% from 2023. Nvidia climbed to the top spot, dethroning both Samsung and Intel, thanks to explosive demand for its GPUs, now powering most AI workloads. Intel now faces a steep climb and is making bold moves to reclaim relevance. Analysts view these steps as painful but necessary to tackle long-standing issues like high costs and sluggish decision-making. “Intel has recognized the need to become more agile,” says Kanisha Chauhan, senior principal analyst at Gartner. “These staff reductions appear to be more than just cost-cutting—they’re aimed at realigning the company with that goal.” Faisal Kawoosa, co-founder of Techarc, agrees: “Intel is trying to reorient itself with the IDM 2.0 strategy — to double down on core strengths and become the world’s No. 2 foundry player by 2030.” These cuts are painful but necessary if Intel is to realign for the future, according to Deepak Gupta, Semiconductor & EDA Expert. Still, the path won’t be easy. “While leadership wants to work more closely with engineers to cut delays and spark momentum, these changes also bring uncertainty and anxiety among employees, which can stifle innovation,” Gartner’s Chauhan said, highlighting the risks that come with such large-scale restructuring. The path ahead for a lost icon Looking beyond cost cuts, Intel’s next chapter will depend on how effectively it can sharpen its focus and embrace disruption. According to Techarc’s Kawoosa, the company must boost efficiency in its core strength — laptop processors — while exploring high-growth areas like connected cars and maintaining momentum in the server market. “Somehow the innovation wheel has stalled,” he noted, adding that Intel might benefit from an inorganic push, possibly through strategic acquisitions of promising startups. According to Gupta, the layoffs reflect “deeper strategic gaps” that have widened over the years, particularly in AI and advanced manufacturing. Gartner remains cautiously optimistic. While Intel still lags in the AI race, Gartner’s Chauhan said Intel’s renewed priorities could start bearing fruit in the coming years if it stays the course. The push to become the world’s second-largest foundry by 2030 is central to Intel’s IDM 2.0 strategy. The layoffs, though painful, are part of this reset. “If Intel sticks to its five-year plan,” Kawoosa said, “the results will start to show.” Intel is rewriting its playbook — moving away from a sprawling, do-it-all model to a sharper focus on what it does best: chip fabrication. By shedding legacy functions and narrowing its scope, the company is betting on agility and specialization. “It’s a big shift,” Kawoosa said, “but maybe it’s exactly what Intel needs to stay relevant—and even thrive—in a future shaped by AI.”

Reliable internet service is no longer a nice-to-have for businesses — it’s a must. Whether for cloud apps, video calls, or running an online store, companies need a steady connection to keep things running smoothly, serve customers, and stay ahead of the competition.  Standard broadband might work for basic needs, but businesses with growing digital demands need something stronger. Dedicated internet access (DIA) provides the reliability, speed, and security that enterprises need to succeed in today’s connected world.  What is dedicated internet access? DIA provides businesses with an exclusive connection that is not shared with other users or organizations. Unlike standard broadband connections, it establishes a direct link between your business and your internet service provider, eliminating the competition for bandwidth that characterizes shared connections.  The fundamental difference between dedicated internet lines and shared connections lies in exclusivity. With a dedicated connection, your business receives consistent bandwidth regardless of network traffic or time of day. In contrast, shared connections distribute bandwidth among multiple users, resulting in potential slowdowns during peak usage times. DIA also typically offers symmetrical speeds, meaning upload and download rates are identical—crucial for businesses that frequently transfer large files or host cloud services. The benefits Dedicated internet access delivers multiple advantages that directly affect business performance and operational efficiency. These benefits extend beyond simple connectivity to enhance your organization’s overall digital capabilities, and include: Enhanced reliability and stability. A dedicated internet line significantly reduces downtime risks, providing businesses with consistent connectivity even during peak usage. Unlike shared network connections that fluctuate based on neighborhood usage, DIA maintains steady performance regardless of external factors. This reliability is particularly valuable for organizations that depend on real-time applications or where internet outages directly translate to lost revenue and productivity.Improved speed and performance. DIA delivers greater performance through guaranteed bandwidth and symmetrical speeds. This means your upload speeds match your download speeds, which can be critical for businesses using video conferencing, VoIP services, or cloud applications. With dedicated internet access, file transfers, virtual meetings, and data backups happen faster and more efficiently, boosting overall productivity and mitigating the frustrating delays that can affect employees and customers.Increased security and privacy. An advantage of dedicated internet connections is the added security. Because you’re not sharing the connection with other businesses or consumers, cybercriminals have fewer opportunities to sneak in. Many DIA providers offer extra security features like DDoS protection, advanced firewalls, and proactive round-the-clock monitoring. This approach provides a safer space for handling sensitive business operations and data transfers.Better support for cloud-based applications. Businesses increasingly rely on cloud-based services, from document storage to customer relationship management. A dedicated business internet line ensures these applications run smoothly with minimal latency. The symmetrical speeds particularly benefit cloud applications that require consistent upload and download capabilities. This reliable performance allows organizations to leverage cloud solutions without connectivity limitations.Cost considerations and return on investment Investing in direct internet access is about more than just the upfront cost; it’s about the overall impact on your business. Although DIA is usually more expensive than standard business broadband, it is important to weigh that against the potential costs of downtime, lost productivity, and missed opportunities.   When calculating return on investment (ROI), think about the boost in productivity, improved customer experiences, and less downtime. For many businesses, the stability and performance that come with DIA directly improve the bottom line, making the higher monthly cost worth it. Plus, DIA often comes with service level agreements (SLAs) that guarantee uptime, offering financial protection if there’s ever a service disruption. Choosing the right DIA provider The right partner is an important decision that can have a big impact on your service quality and business operations. Look closely at factors like uptime guarantees, bandwidth options, scalability, customer support, and security features. A provider should offer clear SLAs that spell out performance expectations and solutions for any service issues. Don’t forget to check their support availability; having 24×7 assistance can be a lifesaver when connectivity problems pop up.  Many dedicated internet providers include big-name telecom companies and specialized business internet providers. For example, Optimum Business offers reliable dedicated Internet with solid uptime guarantees, flexible bandwidth options, and business-focused support to keep your operations running smoothly.  Real-world applications Across industries, organizations leverage dedicated internet access to solve connectivity challenges and improve operations in tangible ways.  For example, financial institutions utilize DIA to ensure secure, uninterrupted transactions and trading activities. Healthcare organizations depend on dedicated connections for reliable telehealth services and secure patient data transfers. E-commerce businesses rely on DIA to maintain website performance during traffic spikes, particularly during high-volume sales periods.  In each case, the consistent performance and reliability of dedicated Internet lines directly support mission-critical business functions. Dedicated internet access gives businesses the speed, reliability, security, and support they need to keep things running smoothly and stay ahead of the competition. Investing in a dedicated internet line pays off with improved productivity, enhanced customer experiences, and fewer disruptions.  As businesses rely more on digital tools, having a solid connection to support those needs is becoming more important for growth and success. Are you getting everything you need from your business Internet? Explore all of Optimum’s Business Internet plans.

The European Commission (EC) has just imposed the first two fines for noncompliance with the Digital Markets Act (DMA). After more than a year of investigation, it has imposed fines of €500 million on Apple and €200 million on Meta. In the first case, the agency says that the iPhone manufacturer imposed restrictions on application developers to inform customers about alternative offers outside the App Store. In this way, according to the EC, “consumers are not able to take full advantage of alternative and cheaper offers.” In the case of Meta, the EC says it should have requested users’ consent to combine their personal data between services, something it has failed to do in its binary ‘Consent or Pay’ advertising model. With this model, introduced in November 2023, which does not fall under the DMA, “Facebook and Instagram users in the European Union (EU) could choose between consenting to the combination of data for personalized advertising or paying a monthly subscription for a service are published.” The agency says also that in 2024 the company introduced a new version of the model with an option that uses less personal data to display ads. This is being evaluated and the EC “continues the dialogue with Meta, requesting the company to provide evidence of the practical impact of this new model.” These decisions fall within a geopolitical framework of negotiations between the United States and Brussels to lower tariffs imposed by Donald Trump on certain products. However, the amount of the fines, per El País, is far from the maximum possible; up to 10% of the company’s annual turnover. Teresa Ribera, Commissioner for Competition, stated, “today’s decisions send a strong and clear message. The DMA is a crucial instrument to unlock potential, choice and growth by ensuring that digital players can operate in competitive and fair markets. Apple and Meta have failed to comply with the DMA by implementing measures that reinforce the dependence of business users and consumers on their platforms. As a result, we have taken firm but balanced action based on clear and predictable rules. All organizations operating in the EU must comply with our laws and respect European values.”

Online threats are prevalent and technology is advancing at a rapid pace, making cybersecurity an urgent requirement in safeguarding the digital assets of small and medium-sized businesses (SMBs).  By protecting vital data, sensitive customer information, and core business operations, SMBs are able to not only defend against cyber-attacks but also harness secure internet connections as a catalyst for growth and competitiveness within the digital marketplace. The benefits of a robust cybersecurity strategy There are multiple advantages for SMBs to have a strong security posture, including: Protection of sensitive business data. By implementing strong cybersecurity measures, SMBs can shield their sensitive information, financial records, and intellectual property from cybercriminals. This fosters trust and reliability among customers and partners. Business continuity. In the face of cyber threats, businesses with secure internet connections can maintain seamless operations, prevent disruptions, maintain productivity, and uphold their reputation in the competitive business environment. Stimulation of innovation and expansion. With a secure cyber environment in place, SMBs can focus on innovation and business development, knowing that their digital infrastructure is protected from cyber threats that could impede their growth. The risks of ignoring cybersecurity SMBs sometimes believe their organizations are not targets of cybercriminals. Yet, attackers have a simple motivation: financial reward. Neglecting cybersecurity exposes SMBs to myriad risks and potential consequences, including: Data breaches. Failure to prioritize cybersecurity leaves SMBs susceptible to data breaches, potentially resulting in the compromise of sensitive business and customer information. This can lead to legal and financial ramifications while undermining client trust and loyalty. Financial losses. Cyber-attacks can inflict severe financial damage, ranging from direct financial theft to costly recovery processes and regulatory fines. The organization’s financial stability and viability can be severely threatened in the aftermath of a successful cyber breach. Reputational damage. A breach of trust resulting from inadequate cybersecurity measures can tarnish reputation. Negative publicity, customer distrust, and damaged relationships with partners and stakeholders can significantly impact the brand’s standing and long-term viability. Regulatory penalties. Non-compliance with data protection regulations and industry standards can expose SMBs to regulatory penalties and legal liabilities. Failing to prioritize cybersecurity can lead to costly repercussions and legal entanglements that may further strain resources and credibility. Future-proof with optimal cybersecurity An optimal cybersecurity approach includes a dynamic set of solutions that are designed to fortify an organization’s digital infrastructure, ensuring robust protection against cyber threats, data breaches, and unauthorized access.  For example, by leveraging innovative technologies such as adaptive firewalls, threat detection, and secure VPNs, SMBs can attain greater resilience and a strong, proactive security posture. In turn, this sets the stage for sustainable growth and innovation in the rapidly evolving digital landscape. It also cultivates a future-ready digital presence, bolstering market competitiveness and fostering customer trust and reliability. With a strategic partnership with a cybersecurity provider such as Optimum Business, SMBs can bolster their resilience, embrace innovation, and establish a secure digital foothold, thereby enhancing their market positioning, fostering customer trust, and navigating the digital future with confidence and readiness. Discover how to enhance your cybersecurity by contacting Optimum Business, and embark on a transformative journey toward a secure and resilient digital future.

Apple has been hammered with a huge €500 million ($570 million) fine by Europe’s antitrust authorities, who have demanded changes in Apple’s business practices that will deliver little significant benefit to consumers. Europe says Apple breached its anti-steering obligation under the Digital Markets Act (DMA), and is forcing changes in the company’s business practices, as well as fines. Apple will appeal In a statement, Apple told Computerworld:  “Today’s announcements are yet another example of the European Commission unfairly targeting Apple in a series of decisions that are bad for the privacy and security of our users, bad for products, and force us to give away our technology for free. We have spent hundreds of thousands of engineering hours and made dozens of changes to comply with this law, none of which our users have asked for. Despite countless meetings, the Commission continues to move the goal posts every step of the way. We will appeal and continue engaging with the Commission in service of our European customers.” Europe’s approach has been uniquely Apple-centric, making demands of the company that are not equally made against its competitors. As part of today’s judgment, Apple is also being forced to open up to third-party app store sales, and to permit developers to let customers know of alternative offers outside Apple’s App Store, steer them to those offers and allow them to make purchases. What Europe wants Apple had attempted to craft an approach to these demands that tried to balance platform security and the costs of building the platforms against what Europe wanted. Under those terms, developers had been asked to agree to certain terms, including payment of a Core Technology Fee. The Commission has rejected Apple’s approach, insisting instead that: Apple must change the business terms through which it enables external app store sales. The Apple Core Technology fee may need to be abolished. Apple must make it easier to set up third-party stores. It must make it easy for consumers to install apps from third-party stores. (I fear this also means it cannot warn customers of the risks of using external stores.) Apple must bring itself into compliance with these demands within 60 days or risk periodic penalty payments. It is to be noted that Meta was also fined today. All it really means In theory, these changes mean apps will be sold through multiple competing stores. That won’t be how things shape up, of course. Some stores will turn out to be malware-infested money traps; others will sell illegal or immoral content; others will show themselves over time to lack standards of customer, privacy, or security support; other developers will use the system to fragment the user experience.  Over time we will see stores fail, fall foul to fraud, or go out of business, leaving no clear path for customer compensation or app longevity. Ultimately there will be one or two surviving stores, as well as Apple, and all three will be owned by corporations, rather than by any up-and-coming European tech firm. That latter isn’t going to happen, and if it does, will only be in specific domains. Through these inevitable evolutions, it will be Apple Support that customers first call for help when things do go wrong, even if responsibility rests with the third-party store. Perhaps this is an improvement, but I don’t see it. This fragmentation will also expose European customers to security and privacy attacks, as some of these stores will not hold the same degree of respect for privacy as Apple. Rewarding good behavior? There is a little good news for Apple in Europe’s judgement, which says the company worked ‘constructively’ with EU regulators on the overall antitrust inquiry. Another strand to this had been an investigation into how Apple enabled user choice on its platforms. Europe has closed that strand of the case, saying Apple has acted to regain compliance. Those actions include giving users in the EU choices of alternative apps and an easy interface through which to make such changes. One thing that isn’t clear is the extent to which Europe is demanding that Apple makes all its new features available to third-party competitors from day one. Ostensibly to ensure competition, Apple had complained that this restriction forced the company to develop for its competitors for free, and that the cost of delivering this would require it to develop for multiple operating systems before launch, slowing innovation and costing a lot of cash. Apple already has 500 developers working on EU compliance, which means Europe becomes an increasingly expensive place to do business. Politically driven nonsense One more thing to highlight is the extent to which Apple has become a punching bag for increasing tension between Europe and the US following the imposition of trade tariffs. In the prelude to today’s announcement, some industry watchers had reported that Europe intended to demand a lower fine from the companies it was about to punish. In the case of Apple, that fine still came down to more than half a billion dollars. Perhaps that is seen as small change in Europe, which has given itself the right to demand up to 10% of a company’s global revenue for offenses against the DMA, but it is unlikely to placate the US administration, which has already warned it will act against nations that act against US tech firms. Apple, which is already being punished by the deteriorating relationship between China and the US, may yet end up seeing further challenges if a tit-for-tat trade war between the US and the EU breaks out. At what point do Europe’s own consumers become casualties of increasing the cost of doing business there? You can follow me on social media! Join me on BlueSky,  LinkedIn, Mastodon, and MeWe. 

A broad coalition of AI experts, economists, legal scholars, and former OpenAI employees is urging state regulators to keep OpenAI’s nonprofit foundation in control of the company. Their concern: that the company’s planned restructuring would abandon its legally mandated nonprofit purpose and place control of artificial general intelligence (AGI) in the hands of private investors. “We write in opposition to OpenAI’s proposed restructuring that would transfer control of the development and deployment of artificial general intelligence (AGI) from a nonprofit charity to a for-profit enterprise.” the coalition wrote in an open letter addressed to the Attorneys General of California and Delaware, who together are the company’s primary regulators. The letter’s signatories include Nobel laureates Daniel Kahneman and Joseph Stiglitz and AI pioneers Geoffrey Hinton and Yoshua Bengio. They argue that the proposed restructuring would violate OpenAI’s Articles of Incorporation, which explicitly state the organization is “not organized for the private gain of any person.” The coalition is urging California Attorney General Rob Bonta and Delaware Attorney General Kathy Jennings to exercise their oversight authority to prevent OpenAI’s proposed restructuring, which they argue would undermine the organization’s original charitable mission. The coalition’s appeal is supported by a separate amicus curiae brief filed by twelve former OpenAI employees in an ongoing federal lawsuit. Together, the letter and brief present a rare, coordinated public challenge to the internal governance of one of the world’s leading AI companies. A legally binding mission OpenAI was created in 2015 as a nonprofit with a single, far-reaching goal: to ensure that AGI benefits all of humanity. Its 2018 Charter outlines principles such as broadly distributed benefits, long-term safety, cooperative development, and technical leadership. These values were designed to steer OpenAI’s work even as it began raising external investment. In 2019, OpenAI adopted a capped-profit model, establishing a limited partnership structure under full control of the nonprofit board. This arrangement, the letter notes, was meant to ensure that AGI development would always remain aligned with the public interest. According to the open letter, the company is now seeking to restructure in a way that would eliminate this charitable governance by allowing private shareholders to assume control of AGI development and deployment. The authors argued that this shift is inconsistent with OpenAI’s charitable purpose and violates both California and Delaware nonprofit law. “As the primary regulators of OpenAI, you currently have the power to protect OpenAI’s charitable purpose on behalf of its beneficiaries, safeguarding the public interest at a potentially pivotal moment in the development of this technology,” the letter said. “Under OpenAI’s proposed restructuring, that would no longer be the case.” Former employees validate governance concerns The amicus brief, filed in April 2025, supports claims made in the open letter by offering firsthand accounts from within OpenAI’s leadership and research teams. The twelve former employees worked at the company from 2018 to 2024 and held roles ranging from research scientists to policy leads. According to the brief, internal operations at OpenAI were built around the Charter. Employee performance reviews included assessments of how individuals advanced the mission, and senior leadership—including CEO Sam Altman—frequently referenced the Charter in strategic decisions. But the brief also reveals a gradual shift in internal dynamics. The former employees claim that key governance principles began to erode as commercial interests grew, culminating in efforts to restructure in ways that would sever nonprofit control. “Without control, the Nonprofit cannot credibly fulfill its Mission and Charter commitments, particularly those relating to broadly distributed benefits and long-term safety,” the brief stated. Transparency and legal accountability urged The coalition’s letter closed with a call for legal action. It urged the Attorneys General to demand full transparency about OpenAI’s current and proposed structures. If OpenAI is no longer operating in line with its nonprofit obligations, the authors argue, the state must act to preserve the public mission. “You currently have the power to protect OpenAI’s charitable purpose on behalf of its beneficiaries, safeguarding the public interest at a potentially pivotal moment in the development of this technology,” the letter said. With AGI development accelerating, the outcome of this governance battle may shape not just OpenAI’s future but the trajectory of AI oversight worldwide. At stake is the principle that technologies capable of reshaping economies, labor, and societies should remain accountable to the public—and not be controlled solely by shareholder interests. Whether legal authorities respond to the coalition’s plea could mark a turning point in how the world manages the power and responsibility of frontier AI development.

We all have bad habits. On a Windows PC, it’s easy to fall into patterns that are “good enough” — they get the job done but may not be the fastest or easiest way to achieve your goals. It’s not that these habits are wrong — it’s that some things are just a bit slow and inefficient. But with a few simple changes, you can transform the way you use your PC, speed up your workflow, and have a better all-around computing experience. Here are the habits I think you should consider breaking — and the transformations I recommend. Want more Windows PC tips? Sign up for my free Windows Intelligence newsletter. I’ll send you free Windows Field Guide downloads as a special welcome bonus! 1. Clicking to open apps from the Start menu Clicking through the Windows Start menu isn’t the best way to open apps. If you find yourself scrolling through the list of all apps and trying to remember the precise name of a particular application shortcut, you’re wasting time. Instead, use search — type the name of an app into the Start menu. Pin applications you use to your taskbar — or at least pin them to the top of the Start menu so they’re easy to access. Consider creating a custom keyboard shortcut for the applications you launch most, too — it’s easy. 2. Digging for folders the slow way in File Explorer It’s easy to find yourself clicking repeatedly through the same folders each time you open File Explorer. Even if you know exactly where to look and are not wondering “Where’d I put that…?” the repeated clicks just aren’t very efficient. Instead, make those important folders easier to access without a lot of clicking. If a folder is important to you, right-click it in File Explorer and select “Pin to Quick Access.” After that, it’ll be easily accessible in File Explorer’s sidebar. Folders you pin in this way will also show up when you right-click File Explorer’s icon on your taskbar thanks to jump lists, and they’ll appear in file Open and Save dialogs throughout Windows — you’ll see the same sidebar at the left side of the dialog window. You can also drag and drop folders right to the left sidebar to pin them. Chris Hoffman / Foundry 3. Hunting down settings screen by screen It happens to me, too: You find yourself slowly clicking through the Settings app, looking for That One Particular SettingTM. Where exactly is it? Even if you remember, you may have to click through a variety of screens, one after another. There’s a better way. Instead, use the search feature to find settings: Just type whatever you’re looking for in the search box in the Settings app. You can also press Windows+I (that’s the letter “i”) to launch the Settings app from anywhere and start typing to search. The Settings app can even help you find other settings-related items, like the Device Manager — even if they aren’t technically in the new Settings app itself. 4. Sticking with the default Start menu and taskbar I’ve seen it over and over: So many people never really customize their taskbar or Start menu. That’s a huge mistake. The default taskbar is often full of pinned apps you don’t need — junk from Microsoft and your PC’s manufacturer — as is the default Start menu. You can make your taskbar and Start menu feel more like your own — and you should. To take control, start by unpinning the taskbar icons you don’t want — right-click each and select “Unpin from taskbar.” In the Start menu, do the same in the pinned items area — right-click something and select “Unpin from Start.” Then, pin the shortcuts you do want there. You can configure more than just which apps appear in your Start menu and taskbar, too. For example, on Windows 11, head to Settings > Personalization > Start to customize exactly what shows up in the Start menu — you may want to hide recommended items, for example. To do the same thing for your taskbar, head to Settings > Personalization > Taskbar to adjust what appears there — perhaps you’d rather not see the Widgets icon, for one thing. You might also want to skip Microsoft’s Start menu entirely and turn to a third-party Start menu replacement or try something like Command Palette. The choice is up to you. Microsoft offers some alternative launchers you may prefer to the Start menu, such as Command Palette.Chris Hoffman / Foundry 5. Manually positioning and resizing windows While you can drag windows around on your screen and resize them by clicking and dragging, that’s not exactly the most efficient way to use a Windows PC. Instead, turn to the built-in Snap feature — which is easy to use with drag and drop, shortcuts like Windows+Left arrow and Windows+Right arrow, or even Windows+Z on Windows 11. But Snap is just one option. If you have a larger monitor or just more advanced needs, Microsoft’s FancyZones PowerToy is an excellent way to take control over how window positioning works on your PC. It can even automatically move windows to their last known zone on screen when you launch them. PowerToys Workspaces is also a quick and easy way to launch a variety of apps from one shortcut, automatically positioning them where you like them. 6. Copy-pasting the tedious way Want to copy paste? That usually means Ctrl+C and Ctrl+V — or the Copy/Paste buttons. But there are better ways. I really do recommend using the clipboard history built into Windows — launch it by pressing Windows+V. Here, you’ll find items you recently copied so you can easily insert them without hunting for them again. It also lets you pin items (text and images) that you frequently paste to save yourself time in the future. Clipboard History is probably one of the most underappreciated features on a modern Windows PC. Chris Hoffman / Foundry The Clipboard History tool also helps you quickly paste something without formatting — something that can be very annoying on Windows. In many apps — like Google Chrome, for example — the Ctrl+Shift+V shortcut will quickly paste without formatting (paste as plaintext), too. For an even more advanced clipboard manager, try Ditto. 7. Capturing screenshots the hard way You don’t have to capture a full-screen screenshot and cut out the bit you want anymore. Windows now offers a much easier way to do this. When you want to share just a small part of your screen, press Windows+Shift+S (or Print Screen) and draw a rectangle around whatever you want to capture. (After pressing the shortcut, you may have to use the options on the bar at the top of the screen to select “Rectangle.” You can then quickly paste that into whatever app you like. 8. Letting your startup apps pick themselves Many applications on Windows add startup tasks — which’ll pop up when you sign in — making the login process take longer, cluttering your system tray, and just wasting system resources if you don’t need them. But Windows lets you quickly control this from a central place. I prefer using the Task Manager for this — right-click an empty spot on your taskbar and select “Task Manager” or press Ctrl+Shift+Esc to launch it. Then, click the “Startup” tab (you may have to click “More details” first on Windows 10). Go through the list of items here and disable any you don’t need — just right-click an item and select “Disable” to stop it from running when you sign in. For more information, right-click an item and select “Search online” — that may help you identify what it does, if you’re not sure. You almost certainly don’t need both Microsoft Copilots running in the background. Chris Hoffman / Foundry 9. Always using the mouse When you aren’t in the habit of using keyboard shortcuts — or you just don’t know them all — it’s easy to use the mouse all the time. But it’s worth learning at least a few powerful keyboard shortcuts, which will gradually boost your PC-using speeds. Start with some basics like Windows+E to launch File Explorer, Ctrl+Shift+Esc to launch Task Manager, and Windows+I (the letter “i”) to launch Settings. And be sure to use Windows+. or Windows+; (that’s a period or semicolon) to launch the emoji panel, too — that’s a useful one. That is just the tip of the iceberg, also —  Windows is packed with keyboard shortcuts, so I recommend thinking about the tasks you perform most frequently and looking up the keyboard shortcuts for those, as well. 10. Rebooting to fix File Explorer, desktop, or taskbar problems A quick reboot can fix a lot of problems on your PC. But that reboot may not be quite so quick if you have to save your work and then reopen all your programs after your PC restarts. Luckily, you can fix lots of little issues right from the Task Manager. Launch it with Ctrl+Shift+Esc. Then, click over to the “Details” list. On Windows 11, search for “Explorer” at the top of the window. On Windows 10, you will have to hunt down “explorer.exe” in the list. Right-click the explorer.exe process and select “Restart.” Windows will restart the Explorer process that handles the desktop, taskbar, and File Explorer windows —this can fix a bunch of issues, and save you from a full reboot. The Task Manager will help you find and fix lots of other issues, too. For example, if an application is using a lot of CPU or memory, find it in the process list and axe it. The “Restart” option only appears for the Explorer.exe process. Chris Hoffman / Foundry Break the patterns and speed up your Windows workflow These tips only scratch the surface of what you can do, and if even one of them inspired you to find a new, quicker way to use your PC, I promise the efficiency upgrade will be worth it. More importantly, I encourage you to identify the things you’re doing on your PC that just aren’t working for you. If you find yourself using a workflow that seems inefficient or just plain annoying, consider looking for a better way. Windows PCs are incredibly flexible tools. For more productivity upgrades, be sure to dig deep into Microsoft’s free PowerToys package. I mentioned FancyZones, Workspaces, and Command Palette above, but there are so many more excellent tools in that package. This is just the start! Sign up for my free Windows Intelligence newsletter today and I’ll send you three new things to try each Friday. Plus, you’ll get free copies of Paul Thurrott’s Windows Field Guides as a welcome gift.

New research shows that many college-educated workers believe their degrees aren’t necessary for their jobs and say they wouldn’t have gone to college if degrees weren’t required for so many roles. In fact, some workers consider their degrees a complete waste of money, according to a survey conducted by The Harris Poll on behalf of job search platform Indeed. The age of the 772 US professionals surveyed plays a major role in their perception of college investment, displaying a generational divide. While only 20% of Baby Boomers consider their degrees a waste of money, that number jumps to 41% for millennials and 51% for Gen Zers. Indeed’s findings line up with other recent surveys. Last year, the Pew Research Center found that only 25% of Americans believe that a four-year college degree is crucial for obtaining a high-paying job in today’s economy. Additionally, 49% think a degree is less important now than it was two decades ago. In October, a USA Today survey found that about 46% of college graduates felt they could have secured their current job without a degree. This sentiment is particularly strong among younger generations, with 52% of millennials and 42% of Gen Z respondents sharing that view. The Indeed survey results highlight a shift toward valuing skills and experience more than traditional qualifications, a trend that has been ongoing for several years. More employers are leaving behind college degree requirements and embracing a skills-based hiring approach that emphasizes strong work backgrounds, certifications, assessments, and endorsements. And soft skills are becoming a key focus of hiring managers, even over hard skills. Large corporations including Boeing, Walmart, and IBM have signed on to varying skills-based employment projects, such as Rework America Alliance, the Business Roundtable’s Multiple Pathways Initiative, and the campaign to Tear the Paper Ceiling, and pledged to implement skills-based practices, according to McKinsey & Co. “They’ve removed degree requirements from certain job postings and have worked with other organizations to help workers progress from lower- to higher-wage jobs,” McKinsey said in a November 2022 report. The lack of a need for a degree is also affecting freelance workers. When considering new hires, 80% of corporate executives prioritize skills over degrees, with half planning to increase freelance hiring this year to fill in for a gap in AI and other skills, according to a recent study by freelance job platform Upwork. Rapidly changing skills and a shrinking wage gap Caroline Ogawa, a director of research in the Gartner’s HR practice, said that as skills evolve rapidly, many college-learned skills become less relevant over time. Nearly two-thirds of job candidates (64%) agree their job skills are constantly changing, and 48% say they’ve had to learn new skills in the past year to stay competitive, according to Gartner’s own research. The top 10 highest paid skills in tech can help workers earn up to 47% more — and the top skill among them is generative artificial intelligence (genAI), according to Indeed and other sources. Upwork’s study showed “unprecedented growth” in demand for specialized AI skills, which have surged 220% year-over-year. To address skills shortages, organizations are prioritizing skills over credentials, allowing them to tap into nontraditional talent. This approach helps employers find candidates with the right skills, even if they don’t have a college degree, Ogawa said. “This shift in focus, combined with the speed of change for skill relevance, has likely impacted those weighing the importance and cost of investing in those credentials,” Ogawa said. While 70% of Indeed’s respondents said their degrees are relevant to their jobs, 36% feel they were a waste, and 60% believe they could do their jobs without them. Those with student debt (41%) are more likely to feel that way, especially Gen Z (68%) and millennials (64%). For decades, a degree meant higher wages, with college graduates earning significantly more than high school graduates. However, the wage gap has recently plateaued, and the “college wage premium” is no longer growing, according to Indeed’s study. And as the labor market tightens as the result of AI adoption and an economic downturn, employers are tightening their job experience requirements with along with a larger applicant pool. For example, the percentage of software development job postings requiring less than a year of experience has declined as overall job postings in the sector decrease. In April 2022, 3.2% of developer job postings required less than a year’s experience. In 2025, that has dropped to 1.2% of employers advertising for positions requiring less than a year’s experience, according to Indeed’s Hiring Lab. “As the class of 2025 prepares to don their caps and tassels, they’re stepping into a labor market marked by extraordinary uncertainty,” Allison Shrivastava, an economist with  the Indeed Hiring Lab wrote in a blog. “While headline job numbers may still appear strong, these figures are both backward-looking and disproportionately buoyed by growth in just a handful of sectors, notably healthcare and social services.” AI changes everything “Undoubtedly, a college education is not the differentiator it once was. But this issue may become moot, as the AI revolution may completely upend higher ed,” said Arthur O’Connor, PhD, head of the data and information science degree programs at CUNY’s School of Professional Studies. Students pay tens of thousands of dollars for a degree based on an outdated divide, said O’Connor, who authored the book Organizing for Generative AI and the Productivity Revolution. Undergrads pick between a broad Bachelor of Arts degree or a skills-focused Bachelor of Science degree. “This is a hopelessly outdated dichotomy, as both sets of disciplines are essential today,” O’Connor said. “Compare this to AI tutors and virtual assistants that offer free or low-cost personalized, self-paced, competency-based learning, tailored to an individual’s learning style, pace, and aptitudes, on any topic of interest,” he said. “In the long run, the real issue is what you learn; not where or how you learn it.” For today’s students, college is just the start of lifelong learning, O’Connor said. To keep up, universities must overhaul costs, tuition, admissions, staffing, and how they teach. If they do, college degrees can remain relevant and useful. “Those who say college education has completely lost its value due to AI are confusing knowledge with understanding. As knowledge becomes commoditized, the importance of understanding AI’s inputs and outputs becomes more valuable,” O’Connor said.

Stop me if you’ve heard this one before: Google’s got a wild new idea about “reinventing” how we silence our phones. It’s happening now, in the current Android 15 era of our favorite operating system. Google eliminated the simple, one-tap system for silencing a device and putting it into Do Not Disturb mode and instead bundled that setting into a perplexing new Modes menu. Now, when you want your phone to zip its metaphorical lip, you’ve gotta tap the Modes tile within your Android Quick Settings panel, then find and activate Do Not Disturb from there. That may seem like a small distinction, but for something as foundational as silencing your frickin’ phone, it’s an extra step that adds up and quickly feels cumbersome — especially if you’re hunting for it multiple times a day. More than that, it’s an extra slice of confusion that creates needless murkiness around what oughta be a simple-as-can-be core phone function. And, oh yes: We’ve been here before. Way back in the Android 5.0 “Lollipop” era of Android, Google similarly sought to revamp the simple silencing function — with a similar sort of setup, even, around overly complicated “modes.” Long story short, it was bewildering and annoying, everyone hated it, and Google eventually backtracked and brought back a simple one-step silencing system. Now, it seems, history is poised to repeat itself again in yet another amusing Android flip-flop. Freshly uncovered code suggests El Googenphlorp has heard the outcry and is actively working to restore a one-tap Android Do Not Disturb toggle that’ll exist alongside the new Modes option. There’s no telling how long it’ll take to show up for most of us, though. And in the meantime — whether you’re using a device with the latest Android version and its ever-fun Modes maze or not — you’ve got some spectacular ways to not only restore a simple Do Not Disturb toggle on your own but also to improve upon it and make it even more effective. Pick whichever path you prefer, and prepare for an instant and immensely practical upgrade to your day-to-day Android-using experience. [Psst: Love shortcuts? My week-long Android Shortcut Supercourse will teach you tons of time-saving tricks for your phone. Sign up now for free!] Android Do Not Disturb shortcut #1: The simple toggle return We’ll start with the simplest possibility of all, and that’s bringing back the one-tap Do Not Disturb toggle in your Android device’s Quick Settings section — without having to wait on Google to make it happen. This one couldn’t be much easier to manage. Just install the free and aptly named Simple DND app from the Play Store. Open it up and follow the prompt to grant the app the permission to manage your phone’s Do Not Disturb state — then tap the Add Tile button on its main screen and confirm that you want to do that. One tap, and boom: Simple DND puts a simple Do Not Disturb tile right back into Android’s Quick Settings for you.JR Raphael, Foundry And that’s it: You should now see a new single-step Do Not Disturb tile right in your Quick Settings, if you swipe down twice (or swipe from the right side of the screen, on certain devices) to open it. Do Not Disturb, right back where it belongs — thanks to Simple DND.JR Raphael, Foundry That, of course, is simply restoring the once-standard Android behavior in a roundabout way. But you can also make it even easier to access your device’s Do Not Disturb mode and silence or unsilence your phone on the fly, if you want. Android Do Not Disturb shortcut #2: The Pixel tap dance If you’re using a reasonably recent Pixel device, you’ve got an incredibly cool and impossibly convenient way to make Do Not Disturb easy to toggle. For this path, you’ll need to go download a handy (and completely free) little app called Mute. Install it, then open ‘er up and follow the prompts to allow the app to modify your system settings. Next, head into your system settings and scroll down until you see the line labeled “System.” Tap that, then tap “Gestures” followed by “Quick Tap to start actions.” Flip the toggle next to “Use Quick Tap” into the on position, then tap the option for “Open app” to select it — and once it’s active, tap the gear-shaped icon alongside it and select “Mute” from the list that pops up. Quick Tap strangely doesn’t support toggling Do Not Disturb out of the box, but the Mute app fixes that.JR Raphael, Foundry Make your way back out of that menu, and now, anytime you double-tap the back of your Pixel, you’ll feel a brief vibration and see a small visual confirmation that Mute has been activated. Your phone’s media and ringer volume will shoot all the way down to zero, and Android’s Do Not Disturb mode will be activated. Double-tap again, and Do Not Disturb will go back off while your volume settings pop back up to their previous levels. You might have to experiment a bit to figure out the exact right spot on the back of your phone where the tapping gesture works most reliably. On most Pixels, it’s higher than you might expect — right beneath the camera module. Once you know where to tap, though, you’ll have a swift ‘n’ satisfyingly simple new way to switch your phone in and out of Do Not Disturb without having to mess with any on-screen menus. As long as the screen is on — even if the device is locked! — all it’ll take is two quick taps to silence or unsilence as needed. Android Do Not Disturb shortcut #3: The notch nudge No Pixel? Or maybe you just aren’t into the whole tap-a-tap tactic? No problem. With the help of a clever and convenient app called Touch the Notch, you can transform that blacked-out space at the top of your screen — y’know, where the front-facing camera lives — into a simple switch for impressively efficient toggling of your device’s Do Not Disturb status. Just snag the app, then open it once to configure it. The first time you do, you’ll be prompted to pay four bucks to bump up to the app’s full version. You may want to do that, for reasons we’ll explore in a moment, but you don’t have to commit right off the bat. If you’d rather try the app out first and see whatcha think, tap the “x” in the upper-left corner of the upgrade offer to skip it. Now, tap “Required permissions” on the app’s main screen, then tap the checkbox next to “Touch the Notch.” With that one-time initialization out of the way, you can scroll down the app’s main screen and review the available options. If you didn’t upgrade to the app’s full version, you’ll be limited to “Single touch” — which isn’t exactly optimal but will get the job done and at the very least let you see what the app can do. Personally, I’d suggest going with “Long touch,” which does require the one-time $4 upgrade but works brilliantly well and is far less likely to be activated accidentally. Tap that line, then find and select “Do Not Disturb” in the list of choices that comes up and make your way back out of that menu. Touch the Notch adds all sorts of interesting possibilities into an otherwise unfunctional area of your phone’s screen.JR Raphael, Foundry Now, for the real magic: At this point, all you’ve gotta do is press and hold that blacked-out spot at the top of your screen (a.k.a. the notch) for a split-second, no matter what else you’ve got goin’ on with your device, and boom: Android’s Do Not Disturb mode will flip on and back off, with every press you make. srcset="https://b2b-contenthub.com/wp-content/uploads/2025/04/android-do-not-disturb-toggle-notch.webp?quality=50&strip=all 800w, https://b2b-contenthub.com/wp-content/uploads/2025/04/android-do-not-disturb-toggle-notch.webp?resize=300%2C23&quality=50&strip=all 300w, https://b2b-contenthub.com/wp-content/uploads/2025/04/android-do-not-disturb-toggle-notch.webp?resize=768%2C59&quality=50&strip=all 768w, https://b2b-contenthub.com/wp-content/uploads/2025/04/android-do-not-disturb-toggle-notch.webp?resize=150%2C11&quality=50&strip=all 150w, https://b2b-contenthub.com/wp-content/uploads/2025/04/android-do-not-disturb-toggle-notch.webp?resize=640%2C49&quality=50&strip=all 640w, https://b2b-contenthub.com/wp-content/uploads/2025/04/android-do-not-disturb-toggle-notch.webp?resize=444%2C34&quality=50&strip=all 444w" width="800" height="61" sizes="(max-width: 800px) 100vw, 800px">Toggle on, toggle off: Touch the Notch’s Do Not Disturb sorcery in action.JR Raphael, Foundry Doesn’t get much easier than that. And you don’t need any special fixes from Google to make any of these paths instantly available. Get even more advanced shortcut knowledge with my free Android Shortcut Supercourse. You’ll learn tons of time-saving tricks!

The ongoing saga of a Google secure browsing project took yet another turn, with the latest change coming just a few days after the company lost an antitrust lawsuit against the US government. Google this week dropped some features from its Privacy Sandbox initiative, which was initiated in 2019 and aims to reduce the invasiveness of third-party cookies in the Chrome browser. The company said it won’t provide a specific prompt that would allow Chrome users to opt in or out of cookies from third parties. That was a central feature of the project that made it easier for users to secure Chrome experiences. The initial objective of Privacy Sandbox in 2019 was to drop third-party ad cookies for better browser security. That idea was dropped last year amid regulatory concerns. Third-party cookies were reinstated, and Google also said it provides “a new experience in Chrome” for users to be able to adjust their third-party cookie choices. This week Google dropped that idea, saying it wouldn’t roll out the so-called “new experience” — a new standalone prompt for third-party cookies for users to opt in or opt out of cookies from third parties. “Users can continue to choose the best option for themselves in Chrome’s Privacy and Security Settings,” wrote Anthony Chavez, vice president for the Privacy Sandbox initiative, in the blog entry. Chavez wrote that the change came amid changes in the regulatory landscape and browsing experiences. Google last week lost a case against the US, which accused the company of violating antitrust law “by monopolizing open-web digital advertising markets.” Google said it would appeal the ruling. The US Department of Justice originally accused Google of “wielding its dominance across digital advertising markets to force more publishers and advertisers to use its products,” according to a press release in 2023. The Electronic Frontier Foundation last year, in a blog entry, criticized Privacy Sandbox, saying it didn’t eliminate online tracking. Instead, online tracking moved from third-party trackers to Google, which tracks the data from Chrome and shares browsing habits with websites and advertisers. “Despite sounding like a feature that protects your privacy, Privacy Sandbox ultimately protects Google’s advertising business,” EFF wrote. Chavez reaffirmed it won’t deprecate third-party cookies. The company will engage with developers, publishers and advertisers to improve the Privacy Sandbox roadmap. Either way, Google’s reversal of Privacy Sandbox plans still points to the company chasing a closed ecosystem that serves the company’s business interests, said Anand Kashyap, CEO and Founder of Fortanix. Advertisers and media companies will use AI in their own secure systems that can analyze data to deliver targeted advertising on the Internet, Kashyap said. Google is also adding a feature in the third quarter this year that will protect the IP address when in secure mode of Chrome browser. Typically users can be located by capturing and tracking the IP address. The feature will anonymize the IP address, and it will be part of the Incognito mode, which creates a temporary browsing window that deletes browsing data on exit, Google said in a blog post on its Privacy Sandbox website. “The feature will be initially available in certain regions, and we plan to expand the availability over time. IP Protection will launch to Chrome Stable no sooner than July 2025,” Google said on the GitHub page for IP Protection. Chrome’s IP address anonymization feature could be important for enterprise users, said Anshel Sag, principal analyst at Moor Insights and Strategy. “It’s clearly Google trying to enhance the security of Incognito mode and more importantly enhance the privacy of a mode that people assume is inherently private,” Sag said. In terms of privacy, Google’s IP Protection feature could block user IP access for third parties but not for their own services or needs, said Alex Matrosov, CEO of security firm Binarly.io. Google is attempting to adopt a similar approach to Apple’s Private Relay feature, which also hides IP addresses and has posed significant challenges for internet advertisers, including Google itself. “In the modern Internet, the word privacy has really transformed into marketing terminology or something else. It’s a nice feature for the Chrome browser to have, but it raises questions about why it was not implemented before,” Matrosov said.

The United Arab Emirates is planning to use AI to review and adjust existing legislation as well as write entirely new laws, reports the Financial Times. The Middle Eastern country is the first in the world to do so. Other countries are currently using AI to streamline various types of work, but not to create entirely new laws. The UAE plans, for example, to use AI to see how laws affect the country’s population and economy by creating a database of laws with public sector data. The expectation is that AI will make local laws 70% faster and ensure they can be updated regularly. It is currently unknown what kind of AI system the country’s authorities will use. “This new legislative system, powered by artificial intelligence, will change the way we create laws and make the process faster and more accurate,” said Sheikh Mohammad bin Rashid Al Maktoum, Emir of Dubai and Prime Minister and Vice President of the United Arab Emirates, according to the country’s state media.

One of the biggest problems with today’s AI models is that they tend to simply make up answers when they don’t know what’s going on, something called hallucinations. You would think that the number of hallucinations would decrease over time, but according to internal tests from Open AI, the opposite is true. The o3 and o4-mini reasoning AI models produce more hallucinations than their predecessors o1, o1-mini, and o3-mini, Techcrunch reports. In one of the tests, the o3 model hallucinated in 33% of responses, compared to 16% for the o1 and 14.8% for the 03-mini. Open AI has no idea why this is the case, but the company’s developers are looking into it and hopefully it will get better in the long run.

Federal prosecutors warned that Google might leverage artificial intelligence to entrench its search monopoly, demanding “strong measures” to prevent the tech giant from extending its market control into the AI era. In the latest phase of the major antitrust trial that began on Monday, government lawyer David Dahlquist argued that Google has built a system where its control of search helps improve its AI products, which then send more users back to Google search, creating a cycle that keeps competitors locked out of both markets. “Now is the time to tell Google and all other monopolists who are out there listening, and they are listening, that there are consequences when you break the antitrust laws,” Department of Justice attorney David Dahlquist told the court Monday, as the remedies phase of the landmark antitrust case against Google began, reported Reuters. The trial, which follows Judge Amit Mehta’s August 2024 ruling that Google illegally maintained a search monopoly, has evolved into a showdown over how deeply the government can intervene in tech markets—particularly as the industry rapidly shifts toward AI-powered services. The AI expansion strategy Evidence presented in court revealed that Google is employing familiar tactics to dominate the emerging AI landscape. The company has negotiated deals paying Samsung to preinstall its Gemini AI app on smartphones, with options to extend the arrangement through 2028. The Justice Department contended that this mirrors the exclusive agreements with device makers that Judge Mehta previously ruled helped Google maintain its search monopoly. The government’s case portrays a self-reinforcing cycle: Google’s search dominance improves its AI products, which in turn drive users back to its search engine. “This court’s remedy should be forward-looking and not ignore what is on the horizon,” Dahlquist emphasized, as per the report. To underscore its focus on AI competition, the DOJ called OpenAI’s product head for ChatGPT, Nick Turley, to testify—signaling the government’s concern about how search and generative AI are converging. “If Google’s conduct is not remedied, it will control much of the internet for the next decade and not just in internet search, but in new technologies like artificial intelligence,” DoJ Assistant Attorney General Gail Slater said in a statement. Remedies that reshape markets The Justice Department is pushing for interventions that would fundamentally alter Google’s business model. Their proposals include requiring Google to divest Chrome, end exclusive default search agreements, and license its search data to competitors. As a last resort, they’ve suggested Google might need to sell its Android operating system if other remedies fail. However, industry analysts question whether some of these remedies align with market realities. Neil Shah, VP for research and partner at Counterpoint Research, believes the Chrome divestiture may miss the bigger picture. “Chrome separation doesn’t impact Google much in the long run as we are moving from browser and app-centric to an Agentic world where search and content access will happen inside the agent app and browser becomes redundant property,” Shah said. “The AI agent itself becomes the search engine.” Sanchit Vir Gogia, chief analyst and CEO of Greyhound Research, expressed similar concerns about focusing too narrowly on Chrome: “Separating Chrome from Google risks destabilizing a global platform that underpins not just web access, but AI discovery itself.” Both analysts said that more attention should be directed toward Google’s control of the Android ecosystem, where default settings and bundled services most effectively limit competition. The DOJ’s proposal for Google to license its search data to competitors faces significant practical challenges, according to experts. “Mandating Google to license its search data may sound like fair market correction, but risks cascading privacy and compliance fallout,” Gogia noted. “Google’s behavioral query logs are rich and sensitive, anonymizing them without destroying contextual utility is technically tenuous.” Shah was more direct: “DOJ’s remedy of Google sharing search results data turns its advertising-led business model as well as tech stack upside down and won’t be practically feasible.” Google’s defense: national security and innovation Google has framed the government’s proposals as threats to the US technological leadership in the global AI race with China — an argument that has gained traction among some industry observers. “We’re in a fiercely competitive global race with China for the next generation of technology leadership, and Google is at the forefront of American companies making scientific and technological breakthroughs,” Lee-Anne Mulholland, Google’s vice president of regulatory affairs, wrote in a blog post. “With new services like ChatGPT (and foreign competitors like DeepSeek) thriving, DOJ’s sweeping remedy proposals are both unnecessary and harmful,” Mulholland added in the post. Gogia acknowledged the complexity of this argument, “Google’s concerns about national security aren’t misplaced. Fragmenting Google’s ecosystem may slow down America’s cohesive AI response to China. But a strong counter opinion to this is that it is a long-overdue correction to embedded defaults that restrict platform access.” Competitors seek a middle ground As the case unfolds, emerging AI search competitors are advocating for more nuanced solutions. Perplexity, an AI-powered search engine, published a blog post arguing that “the remedy isn’t breakup,” but rather increased consumer choice. “When we think about a search product that’s 10X better than 10 blue links, we also think about being a company that works better with OEMs, carriers, and partners of all kinds,” Perplexity wrote in the blog.  “That’s because the only way we (or anyone else) can compete after all the hard work of building a superior product, is to be chosen.” The company has been asked by both the DOJ and Google to share its opinion. This approach aligns with Shah’s analysis: “Maximum DOJ would end up focusing on the exclusive agreements with OEMs and other ecosystem players to make it a somewhat level playing field where users have the choice of the agent.” The remedies trial is expected to conclude May 9, with Judge Mehta’s ruling anticipated in August. Google has already indicated it plans to appeal.

Generative artificial intelligence (genAI) capabilities and features are coming to unified endpoint management (UEM) platforms — in fact, some are already here — and technology and business leaders need to be prepared for the challenges they might face. Some of the leading UEM vendors are weaving AI and genAI features into their platforms. Here are a few examples: ManageEngine has made its in-house AI-based assistant, Zia, an integral part of its UEM solution, Endpoint Central. Through natural-language interactions with the “Ask Zia” chatbot, IT teams can tap into AI-powered insights, intelligent report generation, and AI-enabled remote support. Upcoming features for the platform include genAI-powered management and security automation. GenAI capabilities will be integrated through Ask Zia, and additional features will be aimed at enhancing device performance optimization and security incident management. Microsoft offers Copilot for Windows Autopatch in its Intune UEM product, which enables AI-driven guidance through every update management stage, from planning and deployment tracking to issue identification and remediation. The genAI tool provides actionable insights so teams can keep endpoints secure and up to date with minimal disruption, according to the company. Other available or upcoming Intune features include Copilot assistance for multiple device queries, endpoint privilege management, and policy management. BlackBerry’s mobile threat defense capability for UEM uses AI and machine learning models for scoring apps and URLs to check for malware and malicious sites and phishing incidents. The company says it is evaluating genAI use cases across both servers and apps for inclusion in future releases, with an emphasis on maintaining customer data privacy. A spokesperson declined further comments on these roadmap features or the approximate timeframe of release. Industry watchers also point to improved script generation, natural-language data extraction and analysis, and end-user support as likely applications for genAI in UEM tools. In a large enterprise, a UEM platform might be managing thousands of user devices and other endpoints and tightly tied to security systems, digital employee experience tools, and other enterprise software. Clearly there’s a potential for challenges around security, user experience, and operational efficiency when genAI is embedded in UEM. Preparation is important for success. Computerworld asked three enterprise mobility analysts for their advice on how businesses can take advantage of genAI in UEM tools while still protecting their users, systems, and data. Ask vendors for key information “The most important first step that organizations can take is to fully understand the vendor’s roadmap for genAI features, along with the architecture that will be used to deliver the capabilities,” said Tom Cipolla, senior director and analyst at research firm Gartner. “Surprise releases of genAI are indicative of a failure to prepare and a potentially weak vendor relationship,” Cipolla said. Technology costs are a common concern of organizations, so executives need to keep tabs on how much genAI features cost and whether the added expense is worth it. “Today, most of these capabilities are beta and offered at no cost,” said Andrew Hewitt, principal analyst at Forrester Research. “However, that may not last, as the cost of genAI is high.” Customers should ask vendors for specifics on what they intend to charge for various genAI features in their UEM platforms — and when, he said. Other big issues include cybersecurity and the privacy of corporate data. “GenAI may be utilizing data that is proprietary to the organization, and sending that to a third-party cloud” could be risky, Hewitt said. It’s a good practice to verify with the UEM vendor that data is being processed locally and protected, he said. To that end, UEM customers need to get guarantees from their vendor about security and privacy protections, Hewitt said. It should be stated in the contract that customers’ proprietary data, including their employees’ private data, is encrypted and will not be used in training genAI models. Gartner’s Cipolla also urged IT leaders to ensure that their UEM vendors are making security a priority with genAI. Ideally, genAI features should be provided in a secure way that isolates personal employee and customer data. “Organizations should carefully review the data privacy protection documentation provided by the vendor, specifically looking for cases where the genAI capabilities of the platform use public large language models to fulfill requests,” Cipolla said. Create guardrails Before deploying any forthcoming genAI capabilities in their UEM platforms, companies should take steps to protect their systems and data. For example, they need to put guardrails in place to make sure proprietary data, such as personally identifiable information for employees, is protected. “Organizations need to build AI governance not just for UEM platforms, but also across the digital workplace stack,” Hewitt said. “They should be doing an inventory of where their data currently resides, what protections they have in place for secure authorization, and doing their due diligence around personal or other sensitive information.” IT organizations should start to think about their automation process, Hewitt added. “What types of approvals and authorizations will be necessary to execute automation in the endpoint management stack?” he said. “How will they plan to gain trust and confidence in AI and automation? How should they measure this? Taking an inventory of existing automation processes could help here, as well as doing some testing of genAI on basic use cases.” Testing genAI features should be done in a safe environment prior to rolling them out. “As with any AI solution, organizations should proceed carefully and employ a ‘block, walk, run’ strategy while they gain comfort with the solution and its security,” Cipolla said. Verify, test, and monitor — with humans in charge As genAI features begin to appear in UEM tools, “organizations should ensure that endpoint device management tasks or functions enabled or assisted by AI have similar or better outcomes” than approaches used previously, said Phil Hochmuth, program vice president, enterprise mobility, at research firm IDC. That means keeping a close eye on AI recommendations and actions. “Teams using AI in IT operations for endpoints must be watchful for AI system misinterpretation, partial or incorrect completion of tasks, and other bad outcomes that affect end-user productivity,” Hochmuth said. Enterprises need to be especially mindful of false or inaccurate recommendations from AI, Hewitt said. Administrators need to conduct a “sanity check” on these recommendations before implementing them in their environment. For example, it’s important to confirm that the recommendations are based on recent or real-time data, he said. Cipolla concurred. “Information delivered via genAI can contain inaccuracies and hallucinations — statements that sound factual but are not accurate — resulting from the large language model used to train the AI,” he said. If genAI results are not verified prior to usage, that could result in significant operational impacts, including loss of data, a brand credibility hit, and a degraded digital employee experience, Cipolla said. “For this reason, genAI must be combined with human expertise to validate generated results,” he said. “Prior to implementation of genAI recommendations, ensure that at least one expert human validates the accuracy of the information. Do not use genAI to validate genAI, as different models could share hallucinations.”  To reduce the risk of inaccurate results, Cipolla recommended using a framework similar to common approaches based on the IT Infrastructure Library (ITIL), where proper vetting of IT changes is performed. “Also, prior to implementing any script in a production environment, ensure that testing is performed to validate that there are no unintended side effects. After implementation, carefully monitor the operation of the system for delayed impacts,” Cipolla said. Share what works and build on success Organizations should not fall into the trap of thinking genAI can replace tech employees. “The accuracy of genAI-produced information within tailored use cases, such as digital workplace management tools, will improve quickly. However, genAI will never be able to replace human intuition, empathy, curiosity, experience, and expertise within the digital workplace,” Cipolla said. To prevent potentially catastrophic results, “genAI must be positioned to augment humans and not be seen as an opportunity to replace humans,” Cipolla said. “Human creativity and expertise combined with genAI is a force multiplier that has the potential to yield significant breakthroughs.” To share and collectively improve on positive results, Cipolla recommended that organizations create wiki-style, easily searchable libraries of prompts (and sample result sets) that can be used to identify common successful prompts. “This can be as simple as a shared spreadsheet, a channel in a collaboration tool, or a basic wiki-style website. Enable all employees to contribute, and recognize those employees who exhibit extraordinary creativity in their prompts,” Cipolla said. “Prompt libraries also can be purchased from vendors as a service,” he noted. Here, too, communication with the UEM vendor is important. Most genAI capabilities will have built-in feedback collection mechanisms, where feedback is routed to the vendor for integration into the program, Cipolla said. In this way, genAI successes (and failures) can be used to improve the features in the future.

Huawei is reportedly preparing to ramp up shipments of its new 910C AI chip to Chinese customers as early as next month, a move that could accelerate the fragmentation of global AI infrastructure and challenge US chip dominance in enterprise workloads. Some of the chips have already been shipped, according to a Reuters report, as Chinese AI companies scramble for domestic alternatives to Nvidia’s H20 – a chip that had, until recently, been freely available in the region.

As PC vendors continue to launch multiple variations of AI PCs, the promises of Microsoft Copilot and Copilot+ PC have demonstrated little relevance to enterprises, who are yet to see enough use cases to justify the purchase. However, Lenovo is trying to change that with its own AI assistant, AI Now. Based on Meta’s Llama 3.0, Lenovo AI Now is a small language model that focuses on a limited number of tasks such as document organization and device management, but does it locally, so users don’t have to worry about exposing their data to the outside world, and tasks such as document summaries can be done even on an airplane, without an internet connection. The inspiration for AI Now comes from Copilot’s limited success as a PC assistant. “I think at the launch of Copilot in June last year, there was a promise of more, and they had to pull back. I think it’s going to take some time for us to see wide-scale deployment, especially for some of the features like Recall, which they [Microsoft] showed and then came back off of,” said Tom Butler, VP for worldwide commercial portfolio and product management at Lenovo. “For AI Now, we have used a very focused local model experience. We’re not trying to be cloud. We’re demonstrating just really two things. One is the knowledge assistant, which uses your personal knowledge base; you put specific sets of files, documents into that knowledge base. Then you can run queries, comparisons, summarizations, and work through just that set of documents,” Butler said. The other aspect is that of a PC assistant that the user can instruct to change their PC settings without having to navigate the Settings app, by simply telling it, for example, “turn on dark mode.” Challenges with Copilot+ PC Privacy issues related to Microsoft Recall, which was touted as one of the key use cases for Microsoft Copilot+ PC over a typical AI PC, have put a dent in Microsoft’s positioning, said Udit Singh, vice president, Everest Group. “As such, the business case for Microsoft Copilot+ is weak today. However, this is not uncommon with Microsoft, as it has a tendency to start with a weak business case initially, but then beef up its offering over time,” Singh said. Other experts also struggle to see why enterprises would want to buy an AI PC. “At this stage, AI PCs do not offer enough compelling value to justify large-scale investment, particularly for enterprises with existing AI capabilities in cloud or data center environments,” said Eric Helmer, CTO at Rimini Street. “Many AI workloads can already be effectively managed without requiring a dedicated neural processing unit (NPU) on every employee’s device.” As many enterprises are already cautious with IT spending, initiatives that provide immediate and measurable ROI will remain their priority. Helmer pointed out that investing in AI PCs today could mean paying a premium for capabilities that may soon become accessible through standard hardware and software advancements. “Instead of reacting to vendor-driven cycles, CIOs should assess whether AI PCs align with their broader IT modernization strategies and whether the investment makes sense given their organization’s specific needs,” he said. PC vendors like Lenovo are moving in the right direction by developing on-device AI solutions, arguably the most secure form of genAI, said Himani Reddy, PC research manager at Canalys. “This approach addresses enterprises’ top concern: data privacy,” Reddy argued.“At this stage, the market has limited options, and consumers must choose to either adopt the available solutions or wait for future developments. Unless vendors develop their own personal AI for PCs, like Lenovo’s aggressive push in AI with applications like ‘AI Now’ and HP’s ‘AI Companion,’ Copilot+ PC remains the primary option. Although there are privacy concerns associated with Copilot, it remains a more secure option compared to publicly available generative AI models, which enterprises may be hesitant to adopt.” Move to agentic AI While Lenovo’s entry into the AI PC world has started with a localized AI assistant, the company plans to transform this into a platform that will allow enterprises to choose from multiple LLMs and agentic AI offerings from vendors such as OpenAI, Meta, or even DeepSeek. “When you think of the North Star vision that we have as a company, we have two statements. One is smarter AI for all; we want to deliver this at every price point,” Butler said. The other thing that Lenovo is trying to achieve is to turn the PC into the digital twin of the user. “If each of these devices is operating as our bespoke, unique voice, our digital twin, and I can go ask AI Now, ‘plan my flight to the US,’ and it just goes off and does that for me, that time saving can be immense,” Butler said. To create the digital twin, Lenovo plans to take the agentic AI path. With agentic AI, Butler said, AI Now will move from being merely a personal assistant to a digital twin that can extract the maximum value out of the hardware. Butler said that Lenovo is working with multiple software vendors to integrate more LLMs and AI agents to make the digital twin a reality.

Changes to Intel’s executive management team by its new CEO, after just over a month on the job, is proof of the sense of urgency in the company to act quickly to compete with rivals Nvidia, AMD, and TSMC, an industry analyst said Monday. Mario Morales, group vice president, enabling technologies and semiconductors at IDC, was responding to a recent Reuters report in which, according to a memo it quoted from by new Intel CEO Lip-Bu Tan, “networking chip chief” Sachin Katti has been promoted to the role of “chief technology officer and artificial intelligence chief.” In the memo, Tan stated that the company’s data center and AI chip group, and its personal computer chip group, would report to him directly, not as before to Michelle Johnson Holthaus, who will take on new responsibilities. Reuters reported that Tan wrote, “I want to roll up my sleeves with the engineering and product teams so I can learn what’s needed to strengthen our solutions. As Michelle and I drive this work, we plan to evolve and expand her role, with more details to come in the future.” Morales said the appointment of Katti, who previously was vice president and general manager of the Network and Edge Group (NEX) at Intel, is a good move, because “it tells us that that Intel needs to really rebalance and focus on AI, because that’s where we’re seeing the most rapid growth for the semiconductor market as a whole. That’s probably the area that has the biggest gap in terms of Intel versus its competition.” Scott Bickley, advisory fellow at Info-Tech Research Group,added, “Intel is facing a monumental uphill battle. After completely opting out of the mobile and AI chip design innovation waves, they are now faced with a slumping x86 legacy CPU business coupled with their high-risk strategy to move to a sub-2nm design produced in their fabs.” He pointed out that this “includes new process technologies such as Gate All Around [GAA] processing and backside power delivery that have yet to be proven at scale. It would be somewhat of a miracle to expect Intel to leapfrog both Nvidia and their fab partner, TSMC, in the AI space. Even if these new chips are successful in terms of design and performance, they may be a product in search of a market.” It is possible, said Bickley, that “Intel could seek to be the fab of choice for the hyperscalers if they can bend the cost-for-performance curve, but that is a longshot and likely to only garner a fraction of this market. Intel’s best hope is a paradigm shift to whatever is post-AI chip architecture … they are not in an enviable position.” Morales added that giving more power to Intel engineers is important, since the company has “lost its way,” and the move should help the organization start making the changes that it needs to make. Now, he said, there is a need for Intel to act more quickly, adding that when Tan spoke at the recent Intel Vision conference in Las Vegas, he presented with a “humbleness and integrity that I think resonated with a lot of partners and developers there, but you can see that he needs to make these moves very quickly. He is basically a month into the role and already making management changes, and I think you are going to see more announcements later this week when they announce their earnings.” Morales said he would be not surprised to see more employee attrition, because when it comes to revenue per employee, “it is not as good as the competition.” Alvin Nguyen, senior analyst at Forrester, said that while having a dedicated person in charge of AI development is a “good move,” the proof will come in what Katti is able to deliver as the new CTO. While the fab partnership with TSMC is a “nice cash infusion” for Intel, he said, it is still unclear what Intel will get out of the arrangement, and investors are “not going to be happy until those details come out.” According to Nguyen, the longer this takes, the less latitude the new CEO will be given. “The problem is that they are nowhere near the top of their game in terms of products or fab technology,” he said. “They are behind Nvidia, they are behind AMD in terms of AI, in terms of data center CPUs and in some cases the workplace CPU markets.” He said that the move by Tan to streamline the organization “gives him the ability to see more of what is going on, but the question at this point is, is he overwhelming himself, or does it give him the ability to provide better guidance?”

The US reciprocal tariffs war is beginning to hit the Indian IT industry, with contracts getting delayed as customers adopt a wait-and-watch approach. Some of the country’s top tech services exporters — including Tata Consultancy Services (TCS), Infosys, and Wipro — have begun to show signs of strain, as revealed in their recent quarterly earnings calls. Tata Consultancy Services (TCS), Asia’s largest IT services provider, has flagged growing uncertainty, which it said started in February but has now begun to impact project timelines and client decision-making. K. Krithivasan, CEO and MD at TCS, while announcing the quarterly results, stated that while the overall business environment was positive till February, the company started witnessing some amount of uncertainty since March. This has resulted in some project delays and some ramp-downs. “The Consumer Business Group saw heightened caution and delays in discretionary projects, especially in the US. This was driven by the significant drop in consumer sentiment in February, which preceded changes in global trade and tariffs, creating a domino effect on retail CPG (Consumer Packaged Goods) and TTH (Travel, Transport, and Hospitality) industries,” he said. Wipro is facing a similar heat from the US tariff announcement. While the company started the January 2025 quarter on a positive note, gradually during the quarter, the sentiments turned negative, and the company started witnessing the impact in the US as well as the European market. This is because of the tariff hike and the anticipation around that, which had a cascading impact, Srinivas Pallia, CEO & MD of Wipro, said during the company’s recent analyst call. He also acknowledged that some of the clients in Europe have also slowed down transformation projects, and want to relook at the timelines at this point in time. Infosys, on the other hand, said that its clients have started experiencing tariff pressures, but that hasn’t led to any impact on any existing client discussions. Neil Shah, Vice President Research at Counterpoint Research, said the mounting pressure for onshoring from the US government, beginning with tariffs on goods, will tend to impact IT outsourcing and offshoring – sectors where US firms have long relied on global partners like TCS due to talent gaps, global reach, and cost competitiveness. Experts also said that many companies want to wait out the 90-day tariff hold and reassess their IT spending once the US trade posture becomes clearer. “The biggest challenge for IT budgets is uncertainty.  That being said, projects that are focused on cost reduction, exponential efficiency, and regulatory compliance are still being funded,” said Ray Wang, principal analyst and founder at Constellation Research. The 90-day pause: No relief yet The 90-day pause of tariffs by the White House was meant to give companies some breathing room. However, the reality on the ground is starkly different as it has deepened hesitation among enterprise clients, triggering deal delays, project suspensions, and a slowdown in digital transformation spending. “We were doing a large SAP program, which was very critical for the client, and this was in the consumer sector. And when the client heard about the tariff situation, they were bang in the middle of that, and they put the whole program on pause. Not because they don’t want to do the program, but they wanted to understand, get the certainties of the tariff situation,” said Pallia of Wipro. Even Krithivasan of TCS added that there would be delays in decision-making on discretionary spend if this uncertainty continues. “There are two types of IT services contracts – ‘run the business’ and ‘grow the business.’ ‘Run the business’ will continue while ‘grow the business,’ which is dependent on discretionary spending, will be impacted. Clients might prefer shorter deals for ‘grow the business.’ For ‘run the business,’ clients will expect cost optimization with GenAI, so the contract value of these deals may come down,” said Pareekh Jain,  CEO at EIIRTrend & Pareekh Consulting. Jain said that contrary to the belief that there will be an increase in discretionary spending and more contracts in growth the business with GenAI, but that looks doubtful because of tariff uncertainty. What happens after the tariff pause ends? The biggest question remains: what happens once the 90-day tariff hold ends? If tariffs are reinstated — or even expanded — could it potentially lead to contract renegotiations or client attrition? There can be significant near-term structural shifts in how US firms engage with IT companies, driven by government pressure for onshoring and increasing market volatility. In the long term, the potential for AI to reduce reliance on outsourced IT services can add another layer of transformation, said Shah. Wang of Constellation Research said up to two-thirds of the pending contracts would be in this limbo due to tariff uncertainties. Shah of Counterpoint Research said there is a potential 10-20% impact on future growth opportunities due to the current tariff wars and potential rising inflation in the US, necessitating TCS and other firms to be astute with more flexible engagement models to protect the pipeline until the tariff situation stabilizes. The IT companies are learning to navigate a world where trade policy, not technology, has emerged as the biggest disruptor. Amidst all the uncertainty, TCS is hopeful that FY26 will be better than FY25. Infosys is looking to expand in other geographies, such as Japan, in addition to the work the company has been doing in the US.