When Nintendo announced the Switch 2 would cost $450, my initial reaction was disappointment. “Why does it cost so much more?” I thought to myself. “Why does Japan get it cheaper?” my brain jealously added, once I learned that Nintendo would sell a Japan-only model for the equivalent of just $333.It felt like Nintendo was about to overcharge the entire rest of the world for a modest improvement to its original $300 console, one that doesn’t come with an OLED display or anti-drift magnetic sticks. Surely it can’t cost Nintendo that much more to make, especially seeing how it’s selling the exact same hardware for so much “less” in Japan? But while Nintendo might be charging more than I’d like to spend, particularly with its $80 games and its button that makes you pay extra, I no longer think the company’s being distinctly unfair to gamers outside of Japan. The $450 price makes more sense when you consider what’s happened to the dollar and the yen. The original Switch still costs around 33,000 yen in Japan. Image: NintendoSince its March 3rd, 2017, debut, Nintendo basically hasn’t changed the price of the original Switch in either the US or Japan. The portable console cost $300 USD or ¥32,378 in 2017; it costs the same $300 USD or a very slightly higher ¥32,978 today, eight years later. But the US economy and the Japanese economy have not been the same since 2017. Inflation and exchange rates have drastically changed since then.On the one hand, the US dollar now buys less in the US than it did in 2017. Dramatic US inflation means that an original $300 Switch would cost nearly $400 if introduced today. The same is not true in Japan, which has historically had low inflation, or even deflation, until fairly recently. A ¥32,378 Switch would cost more like ¥36,000 today if Nintendo had followed the trend.On the other hand, US dollars are now worth dramatically more in Japan because the Japanese yen is remarkably weak. Compare: on January 13th, 2017, the day Nintendo announced the original Switch’s price, you could only get 114 yen per US dollar. On April 2nd, 2025, the day Nintendo announced the Switch 2, each dollar was worth almost 150 yen. Each dollar buys 30 percent more in Japan than it did back then.Mountain view or no, the Fujikawaguchiko Lawson still attracted crowds when we walked past two weeks ago. Photo by Tom MerrittJapan is experiencing more tourism than ever before as a result of that weak yen, to the point some particularly scenic locations have become overrun — it was international news when a small town near Mount Fuji erected a barrier to keep people from blocking the streets while duplicating an iconic Instagram shot of the mountain merged with a convenience store. I just came back from a two-week trip to Japan, and I’ve never eaten so well for so little cash. But Japanese residents can’t say the same. Not only is Japan starting to experience inflated food prices, but their wages reportedly aren’t keeping pace. It’s not fair to say, “Oh, Japan only has to pay $330 for the Switch 2 while we pay $450,” because both the Japanese and international versions cost almost exactly 50 percent more than Nintendo charges for an original Switch. Going from $300 to $450 is a 50 percent increase; going from the Switch’s ¥32,978 to the Japan-only Switch 2’s ¥49,980 is 51.6 percent. That suggests Nintendo might be telling the truth that the price was not designed to preempt tariffs.You really don’t need a convenience store to take a lovely pic of Mt. Fuji! Photo by Sean Hollister / The VergeInstead, the disparity we see is likely because the dollar goes so much further right now, and because the yen does not. As one Tokyo-based game streamer put it: “Salaries and cost of living hasn’t changed at all here so 49,980 feels like 450 usd.” A fun thought experiment: if exchange rates were the same as they were in 2017, the Japan-only Switch 2 would cost $434 in US money. It would cost $450 USD there if the dollar sinks to 111 yen or below, which it did during large portions of 2018, 2019, 2020, and 2021.If Japan-based Nintendo decided that the right price for the Switch 2 is 50 percent more than the original Switch, it does kind of follow that the US model would cost 50 percent more, too.Even the region lock makes some sense, as much as I hate the idea. If Nintendo didn’t lock it down, foreigners could take advantage of the weak yen to buy and scalp the Japanese versions all over the world, similar to what’s already reportedly happening with Japan’s retro game stores.I couldn’t find a Mother 3 cart in my city trips — one clerk laughed when I asked — but I eventually found this one in the suburbs. Photo by Sean Hollister / The VergeBut can Nintendo justify charging 50 percent more for a new handheld eight years later? That’s where I’m less sure. While US inflation might help justify a $400 Switch (again, the $300 original Switch would cost nearly $400 in today’s money) or even a $80 game, I think it’s more that Nintendo can afford to get away with a $450 Switch because real competition is slim.Sony and Microsoft won’t have true handheld PlayStations and Xboxes for years to come, and handheld PC makers are still struggling to beat the Steam Deck, let alone the Switch. (Nintendo can sometimes sell more Switches in a month than handheld gaming PCs sell in a year.) And while you can currently buy a new PS5 in the US for less than a Switch 2, that isn’t true in Japan — after three price hikes, the cheapest PS5 Digital Edition now costs ¥72,980, 46 percent more than the new Japan-only Switch, and Sony’s pricey PS5 Pro costs a staggering ¥119,980 there. RelatedEven Japan’s cheapest Steam Deck costs 20 percent more than a Switch 2, at ¥59,800 for the 256GB LCD model, so the Switch really and truly stands alone. Perhaps that’s why the gamer vibe in Japan is reportedly quite positive despite a 50 percent higher price. “The general consensus among Japanese gamers in response to this has been “Thank you Nintendo, please become the ruling party of our country next,” writes Automaton.The vibe definitely isn’t as rosy in the US, but at least the price now makes some sort of sense to me. And now that President Donald Trump has given companies like Nintendo a 90-day reprieve from many of his tariffs, a period of time that covers the Nintendo Switch 2’s June 5th launch, Nintendo may be able to have its cake and eat it, too. Nintendo can launch with its promised price and blame Trump for any later increase. Perhaps today’s $450 will look generous by comparison. See More:

As AI adoption increases in digital infrastructure, enterprises and developers face mounting pressure to balance computational costs with performance, scalability, and adaptability. The rapid advancement of large language models (LLMs) has opened new frontiers in natural language understanding, reasoning, and conversational AI. Still, their sheer size and complexity often introduce inefficiencies that inhibit deployment at scale. In this dynamic landscape, the question remains: Can AI architectures evolve to sustain high performance without ballooning compute overhead or financial costs? Enter the next chapter in NVIDIA’s innovation saga, a solution that seeks to optimize this tradeoff while expanding AI’s functional boundaries. NVIDIA released the Llama-3.1-Nemotron-Ultra-253B-v1, a 253-billion parameter language model representing a significant leap in reasoning capabilities, architecture efficiency, and production readiness. This model is part of the broader Llama Nemotron Collection and is directly derived from Meta’s Llama-3.1-405B-Instruct architecture. The two other small models, a part of this series, are Llama-3.1-Nemotron-Nano-8B-v1 and Llama-3.3-Nemotron-Super-49B-v1. Designed for commercial and enterprise use, Nemotron Ultra is engineered to support tasks ranging from tool use and retrieval-augmented generation (RAG) to multi-turn dialogue and complex instruction-following. The model’s core is a dense decoder-only transformer structure tuned using a specialized Neural Architecture Search (NAS) algorithm. Unlike traditional transformer models, the architecture employs non-repetitive blocks and various optimization strategies. Among these innovations is the skip attention mechanism, where attention modules in certain layers are either skipped entirely or replaced with simpler linear layers. Also, the Feedforward Network (FFN) Fusion technique merges sequences of FFNs into fewer, wider layers, significantly reducing inference time while maintaining performance. This finely tuned model supports a 128K token context window, allowing it to ingest and reason over extended textual inputs, making it suitable for advanced RAG systems and multi-document analysis. Moreover, Nemotron Ultra fits inference workloads onto a single 8xH100 node, which marks a milestone in deployment efficiency. Such compact inference capability dramatically reduces data center costs and enhances accessibility for enterprise developers. NVIDIA’s rigorous multi-phase post-training process includes supervised fine-tuning on tasks like code generation, math, chat, reasoning, and tool calling. This is followed by reinforcement learning (RL) using Group Relative Policy Optimization (GRPO), an algorithm tailored to fine-tune the model’s instruction-following and conversation capabilities. These additional training layers ensure that the model performs well on benchmarks and aligns with human preferences during interactive sessions. Built with production readiness in mind, Nemotron Ultra is governed by the NVIDIA Open Model License. Its release has been accompanied by other sibling models in the same family, including Llama-3.1-Nemotron-Nano-8B-v1 and Llama-3.3-Nemotron-Super-49B-v1. The release window, between November 2024 and April 2025, ensured the model leveraged training data up until the end of 2023, making it relatively up-to-date in its knowledge and context. Some of the Key Takeaways from the release of Llama-3.1-Nemotron-Ultra-253B-v1 include: Efficiency-First Design: Using NAS and FFN fusion, NVIDIA reduced model complexity without compromising accuracy, achieving superior latency and throughput. 128K Token Context Length: The model can process large documents simultaneously, boosting RAG and long-context comprehension capabilities. Ready for Enterprise: The model is ideal for commercial chatbots and AI agent systems because it is easy to deploy on an 8xH100 node and follows instructions well. Advanced Fine-Tuning: RL with GRPO and supervised training across multiple disciplines ensures a balance between reasoning strength and chat alignment. Open Licensing: The NVIDIA Open Model License supports flexible deployment, while community licensing encourages collaborative adoption. Check out the Model on Hugging Face. All credit for this research goes to the researchers of this project. Also, feel free to follow us on Twitter and don’t forget to join our 85k+ ML SubReddit. Asif RazzaqWebsite |  + postsBioAsif Razzaq is the CEO of Marktechpost Media Inc.. As a visionary entrepreneur and engineer, Asif is committed to harnessing the potential of Artificial Intelligence for social good. His most recent endeavor is the launch of an Artificial Intelligence Media Platform, Marktechpost, which stands out for its in-depth coverage of machine learning and deep learning news that is both technically sound and easily understandable by a wide audience. The platform boasts of over 2 million monthly views, illustrating its popularity among audiences.Asif Razzaqhttps://www.marktechpost.com/author/6flvq/Together AI Released DeepCoder-14B-Preview: A Fully Open-Source Code Reasoning Model That Rivals o3-Mini With Just 14B ParametersAsif Razzaqhttps://www.marktechpost.com/author/6flvq/Boson AI Introduces Higgs Audio Understanding and Higgs Audio Generation: An Advanced AI Solution with Real-Time Audio Reasoning and Expressive Speech Synthesis for Enterprise ApplicationsAsif Razzaqhttps://www.marktechpost.com/author/6flvq/Interview with Hamza Tahir: Co-founder and CTO of ZenMLAsif Razzaqhttps://www.marktechpost.com/author/6flvq/OpenAI Open Sources BrowseComp: A New Benchmark for Measuring the Ability for AI Agents to Browse the Web

I don’t usually get excited about Magic: The Gathering deals unless they involve a substantial discount or a chance to pick up chase cards without pawning my fetch lands. But the current Best Buy Deal of the Day has me genuinely interested, and not just because I’m weak in the face of shiny foils and stitched-edge playmats. As with all DotD discounts, this is limited just to today as well, so pick up the cheap boosters while you can. Let's dig into what's available, and what I'd recommend picking up.MTG Discounts On Foundations, Aetherdrift, and More Wizards of The Coast - Magic: The Gathering Foundations BundleWizards of The Coast - Magic: The Gathering Foundations Jumpstart Booster 2-Pack Wizards of The Coast - Magic: The Gathering Ravnica Remastered Draft Booster SleeveWizards of The Coast - Magic: The Gathering Aetherdrift BundleWizards of The Coast - Magic: The Gathering Aetherdrift Collector BoosterWizards of The Coast - Magic: The Gathering Aetherdrift Commander Deck - Living EnergyWizards of The Coast - Magic: The Gathering Fallout Commander Deck - Science!Ultra PRO - Mana 8 - Stitched Edge Playmat - Plains for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Mountain for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Forest for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Island for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Lotus for Magic: The GatheringLet’s talk Foundations. This set isn’t trying to wow anyone with lore about a haunted mansion or a squirrel-led uprising. Instead, it focuses on being playable, readable, and shockingly welcoming. I think that’s a big deal, especially since we’re about to get Marvel and Final Fantasy collaborations through Universes Beyond. New players are going to get dropped into the multiverse with zero context, so a solid on-ramp like Foundations is going to be essential.Magic: The Gathering Foundations BundleWizards of The Coast - Magic: The Gathering Foundations BundleIf you’ve never touched a deck in your life, the Foundations Beginner Box is a better teacher than half the humans I’ve seen try to explain Magic at a kitchen table. Pre-constructed decks, visual aids, and a guided tutorial make it absurdly approachable. t. There’s also the Foundations Bundle, which is going for $42.49. That's under market price and loaded with boosters and accessories. It’s great value, especially since Foundations is going to stick around in Standard for five years, giving you time to figure out what the heck a stack is.Magic: The Gathering Foundations Jumpstart Booster 2-Pack Wizards of The Coast - Magic: The Gathering Foundations Jumpstart Booster 2-Pack And then there's Aetherdrift, which is for people who read the card Blasphemous Act and said, “but what if we did more?” The Aetherdrift Bundle is currently $34.99, which is a solid $16 off its usual price. It includes some of the more unhinged red spells and flashy threats to come out of Standard lately, including Marauding Mako (which is basically Magic's version of a landshark with anger issues) and Momentum Breaker, a mono-black tool that's found a home in aggressive Pixie decks. Yes, apparently that’s a thing now.Magic: The Gathering Ravnica Remastered Draft Booster SleeveWizards of The Coast - Magic: The Gathering Ravnica Remastered Draft Booster SleeveFoundations also includes some pretty desirable chase cards if you're cracking packs. I’m looking at you, Rise of the Dark Realms with the new Borderless treatment, and Sire of Seven Deaths, the Eldrazi with seven keywords that feels like it was printed specifically to cause someone a rules headache. And don’t sleep on Banner of Kinship, which casually turns your Goblin deck into a math test your opponent definitely didn’t study for.Magic: The Gathering Aetherdrift BundleWizards of The Coast - Magic: The Gathering Aetherdrift BundleYes, I noticed Ravnica Remastered is part of this sale. You can grab a Ravnica Remastered Draft Booster Sleeve for $4.99 (down from $6.99), which is actually one of the cheapest ways to chase after format staples like Crypt Ghast, Spark Double, and Divine Visitation. I think that’s a great price for anyone looking to bulk open product or toss a few into a sealed night with friends.Magic: The Gathering Aetherdrift Collector BoosterWizards of The Coast - Magic: The Gathering Aetherdrift Collector BoosterPersonally, I love Ravnica sets. I’ve always been a sucker for guild mechanics, and Ravnica Remastered is basically Wizards saying, “Let’s throw all the good stuff into one nostalgic blender.” You get a mix of old-school Commander hits like Cyclonic Rift, weird-but-powerful tech like Cloudstone Curio, and alternate art versions that will either delight you or make you question whether you’re playing Magic or flipping through concept art for a lost anime seriesMagic: The Gathering Aetherdrift Commander Deck - Living EnergyWizards of The Coast - Magic: The Gathering Aetherdrift Commander Deck - Living EnergyI picked up a few sleeves myself, partly for the shot at Anime Borderless Bruvac, and partly because I have no impulse control when it comes to reprint sets. You don’t need to justify $5 packs when they include Shock Lands and cards with infinite combo potential.Magic: The Gathering Fallout Commander Deck - Science!Wizards of The Coast - Magic: The Gathering Fallout Commander Deck - Science!If you're already building for Commander or just pretending your collection isn’t entirely impulse buys, the Fallout Commander Deck – Science! is on sale for $33.99. That’s $26 off retail, and includes one of the better face Commanders to come from Universes Beyond: Dogmeat, Ever Loyal, who floods the board with Junk tokens if you build around equipment and auras. In my opinion, this deck is an underrated engine, and you get to play cards like Radstorm and Feral Ghoul, which sound like B-side punk albums but are surprisingly effective in-game.MTG: Ultra Pro Mana 8 Stitched Edge PlaymatsUltra PRO - Mana 8 - Stitched Edge Playmat - Plains for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Mountain for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Forest for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Island for Magic: The GatheringUltra PRO - Mana 8 - Stitched Edge Playmat - Lotus for Magic: The GatheringI don’t usually care about playmats, but these Mana 8 stitched-edge ones are kinda gorgeous. At $16.49, they’re dramatic in all the right ways without costing you the same as a new commander deck. I think the Lotus one is low-key perfection.MTG: First Look at Cards From Tarkir: DragonstormWhile Magic: The Gathering’s big crossovers with things like Final Fantasy and Spider-Man might be stealing the spotlight lately, the next set in the line-up is already storming toward us. Tarkir: Dragonstorm returns to Magic’s own plane of Tarkir, and we’ve got an exclusive first look at a five cards that I expect we’ll be seeing a lot of at the draft table next month. (The expansion is available to preorder now on Amazon).Flip through the gallery below to see all five cards, and read on for a word from Wizards of the Coast about their design and the set as a whole.Magic: The Gathering - 5 New Cards from Tarkir: DragonstormChristian Wait is a contributing freelancer for IGN covering everything collectable and deals. Christian has over 7 years of experience in the Gaming and Tech industry with bylines at Mashable and Pocket-Tactics. Christian also makes hand-painted collectibles for Saber Miniatures. Christian is also the author of "Pokemon Ultimate Unofficial Gaming Guide by GamesWarrior". Find Christian on X @ChrisReggieWait.

It is an image that rocked the internet, from online scientific publications to pop culture websites like our own. An animal that Colossal Biosciences (and for that matter the cover of Time magazine) declared to be a dire wolf is now a snow-white puppy with huge feet and olive eyes—and it’s howling. According to Ben Lamm, the CEO and co-founder of Colossal, the sight even brought investor Peter Jackson to tears. After all, this is the first time anyone’s heard a dire wolf cry in about 10,000 years, right? Now after so much time, the Pleistocene predator is back. The sheer audacity of such a broad assertion is also why some are applying intense, if not occasionally hostile, skepticism toward the idea. For while the headlines emphasize the scientific marvel of the creation of three wolves named Romulus, Remus, and Khaleesi, via what Colossal champions as de-extinction innovations, a closer look at the science adds an important caveat: Colossal is pursuing what it calls “functional de-extinction.” As per the Dallas-based biotech company’s website, functional de-extinction is “the process of generating an organism that both resembles and is genetically similar to an extinct species by resurrecting its lost lineage of core genes; engineering natural resistances; and enhancing adaptability that will allow it to thrive in today’s environment of climate change, dwindling resources, disease and human interference.” The difference between directly cloning dire wolf DNA—which is functionally impossible at the moment without a living cell—and genetically engineering the DNA of a living creature to match the genome of its extinct, distant relative is a fine, fine line. However, if Colossal’s assertions stand up to peer review, then they were successful in at least the first goal of their stated mission: generating an organism that both resembles and is genetically similar to an extinct species. While the company’s scientists prepare a paper explaining their results, it currently seems they were able to extract historic genetic detail about the dire wolf by pulling ancient DNA from a 13,000-year-old dire wolf tooth discovered in Ohio and a 72,000-year-old skull in Idaho. This would also mark the first successful sequencing of a dire wolf’s genome. Furthermore, Colossal was able to create the embryos that became their three wolves by successfully editing 20 genes found in a gray wolf to match those Colossal identified as divergent between the dire wolf and gray wolf—which are otherwise more than 99 percent identical. However, even then only 15 of Colossal’s multiplex gene edits are an exact match with the ancient dire wolf; the other five are similar mutations found in modern gray wolves and dogs, because the exact mutations discovered in Colossal’s reconstructed dire wolf genome could reportedly cause blindness and deafness in a gray wolf, whose cells are still the basis for Colossal’s breakthrough. These finer details is what has caused many to be acutely suspicious of calling Romulus, Remus, and Khaleesi dire wolves. University of Maine paleoecologist Jacquelyn Gill posted on Bluesky, “Making genetically modified animals that are cosplaying as extinct species is not de-extinction… They are selling you de-extinction but are creating genetically modified animals with some traits of an unrelated extinct species.” Meanwhile even scientists more receptive to the possibilities presented by Colossal’s research are wary of calling these three animals full-on dire wolves. In an interview with the New York Times, Cornell University geneticist Adam Boyko expressed enthusiasm toward learning something about ancient dire wolves from observing Romulus, Remus, and Khaleesi—for instance discovering that dire wolves were not only all-white but featured thick manes around their necks and a fox-like tail—but still would not consider them truly resurrected versions of the extinct species. How many other genes, Boyko speculated, in a species that walked the earth for more than a hundred thousand years could there be differentiating ancient dire wolves from modern gray wolves? ““We don’t know what that number is. It could be 20 or it could be 2,000.” Ironically for a company whose research has invited plenty of comparisons to Michael Crichton’s Jurassic Park—including by us—this debate resembles a crucial detail that the notoriously cynical author buried in his book, and which noted showman Steven Spielberg left out of the more famous 1993 film adaptation. In the novel, the head of Jurassic Park’s laboratory, which brought back dinosaurs via blood preserved by mosquitos frozen in amber, muses to himself that what they created are not actually dinosaurs. They used the genetic code of modern-day amphibians to fill in gaps in dinosaur genomes. They also reengineered the creatures to attempt to make them more docile. “I don’t think we should kid ourselves,” the character finally says. “We haven’t recreated the past here. The past is gone; it can never be recreated. What we’ve done here is reconstruct the past—or at least a version of the past.” It’s a timely comparison to the basic idea of the de-extinction project, and one which we raised with Colossal head Lamm the first time we interviewed him at SXSW when discussing their breakthrough with woolly mice. To be clear, Colossal is not filling in gaps of genetic code with modern animals like the dinosaurs in Crichton’s Jurassic Park. With their dire wolves they manipulated a modern genetic code that was similar to the dire wolf to be what it claims is a full-on or near-enough match. (Additionally, Lamm has assured us that there is no way to bring back a dinosaur, despite what imaginative authors say about amber and mosquitos.) Join our mailing list Get the best of Den of Geek delivered right to your inbox! Yet while discussing the company’s stated goal to functionally de-extinct the woolly mammoth by 2028 through the use of Asian elephant DNA, Lamm made an argument to us that seems prescient to the current one over Colossal’s dire wolves. “I think it’s important that people understand the difference between what’s possible and what’s not,” Lamm said at the time regarding the woolly mammoth project and the difference between functional de-extinction and cloning a living cell. But he went on to compare the debate to a person who learned on 23andMe that they have Neanderthal DNA in their ancestry. Indeed, non-African humans can have anywhere between one and four percent Neanderthal DNA in their genetics (it is closer to zero percent among African populations).  “Would you say that you’re a hybrid or do you say you’re a Neanderthal-hybrid with Homo sapien, or do you say that you’re an evolved Neanderthal, because they did come first, or do you say that you’re a Homo sapien?” Lamm argued. The biotech entrepreneur also pointed out that humans love classifying, quantifying, and compartmentalizing data with clear lines, but compares de-extinction technology to the evolutionary process of speciation, which in nature can occur over tens of thousands of years. “Speciation is more like a river,” Lamm said. “All of these species evolve from each other, and we’re all hybrids of a hybrid of a hybrid. So I wouldn’t say it’s correct to classify our mammoth as an Asian elephant or hybrid…  It isn’t a cold-adapted or allele-adapted Asian elephant, which is what some people love to say about our work. We think that if it solves the functional aspects of a mammoth, if it has ancient DNA from a mammoth, and if it has the lost genes to a mammoth, we just call that a mammoth. If people want to call it ‘mammoth 2.0,’ they can. Or if people want to say 40 words to describe it, they can too.” Lamm would likely make the same argument about the three snow-white wolves his company created, two of which are already the same size or bigger than modern gray wolves despite not even being halfway grown to adulthood. Whether that counts as a dire wolf really seems to come down to what your definition of a dire wolf might be, and folks far more learned, and with the PhDs to prove it, will assuredly argue that point for many years to come as we step further into a future where the possibilities of biotechnology are only beginning to be unlocked. With that said, whatever you are wont to call Romulus, Remus, and Khaleesi, they do provide a fascinating window into better understanding the original dire wolves. Already their mere existence has revealed hitherto unknown secrets about the ancient creature’s appearance and fur. There is opportunity in this. Perhaps more functionally important though is that their existence also acts as a kind of neon-lit billboard for how such technology can be used to save currently endangered species. Buried beneath all the headlines, including the one at the top of this article, is that Colossal also used the same research that created their dire wolves to successfully clone four red wolves through a combination of extracting red wolf DNA from coyote-wolf hybrids found in the North American southeast, as well as an ancient and, until now, dead line of red wolf DNA. “If [the U.S. government wants] me to make a thousand red wolves that are genetically diverse, we could do that for them,” Lamm said.  “We’ll make all of our technologies available for free for conservation. It’s part of our mandate.” If biodiversity can be restored to a species humans have hunted to the point where there are only between an estimated 17 and 19 red wolves left in the wild, that might be the real miracle of this technology.

Apple last year introduced a beta version of Apple Maps on the web, which allowed users of other platforms to access the company’s mapping service for the first time. However, Apple Maps on the web was labeled as “beta” — until now. And as the platform is no longer beta, it has also gained support for mobile devices. Apple Maps web app now available for mobile devices With the change, Apple Maps on the web is now available at maps.apple.com, instead of beta.apple.maps.com. But the biggest surprise is the support for mobile platforms. Whereas previously the web version of Apple Maps was only available for desktop and tablet web browsers, it now works on iPhone and Android phones too. While the news won’t make a difference to the vast majority of iOS users, Android users can now choose to use Apple Maps if they want. The web app lets users explore the map, search for businesses and points of interest, find directions and view Guides. Features such as the transit map and 3D buildings remain unavailable. There’s also no option to sign in with an Apple Account to access saved places and custom Guides. A few months ago, Apple updated the Maps web app with support for the Look Around feature, which lets users explore their surroundings with 360-degree panoramic views. Where available, Look Around can be accessed by tapping or clicking on the binoculars icon in Apple Maps. Apple says it will continue to add new features and support for more platforms to the Apple Maps web app. H/T: Carlos! Gadgets I recommend: Add 9to5Mac to your Google News feed.  FTC: We use income earning auto affiliate links. More.You’re reading 9to5Mac — experts who break news about Apple and its surrounding ecosystem, day after day. Be sure to check out our homepage for all the latest news, and follow 9to5Mac on Twitter, Facebook, and LinkedIn to stay in the loop. Don’t know where to start? Check out our exclusive stories, reviews, how-tos, and subscribe to our YouTube channel

Apr 11, 2025The Hacker NewsCybercrime / Security Breach What are IABs? Initial Access Brokers (IABs) specialize in gaining unauthorized entry into computer systems and networks, then selling that access to other cybercriminals. This division of labor allows IABs to concentrate on their core expertise: exploiting vulnerabilities through methods like social engineering and brute-force attacks. By selling access, they significantly mitigate the risks associated with directly executing ransomware attacks or other complex operations. Instead, they capitalize on their skill in breaching networks, effectively streamlining the attack process for their clients. This business model enables IABs to operate with a lower profile and reduced risk, while still profiting from their technical skills. Operating primarily on dark web forums and underground markets, IABs can function independently or as part of larger organizations like Ransomware-as-a-Service (RaaS) gangs. They act as a crucial link in the cybercrime ecosystem, providing the initial foothold needed for ransomware gangs, data thieves, and other malicious actors to carry out their operations. The pricing of their services is dependent on the target's size, the level of access granted, and the perceived value of the compromised system, typically conducted within the dark web. Why are IABs gaining steam? The rising prominence of Initial Access Brokers (IABs) is directly tied to their ability to streamline and accelerate ransomware operations, particularly Ransomware-as-a-Service (RaaS) schemes. By handling the complex task of initial network infiltration, IABs allow ransomware groups to focus solely on data encryption and extortion, effectively scaling their attack capabilities. This efficiency is further amplified by the growing trend of IABs working directly for RaaS affiliates, enabling near-instantaneous attacks upon access procurement, eliminating the time-consuming process of establishing a foothold. This symbiotic relationship benefits both sides. RaaS groups gain speed and efficiency, while IABs secure a consistent stream of work, often bypassing the need for public advertising on dark web forums. This reduced visibility provides a layer of protection from law enforcement scrutiny, as their activities are less exposed compared to those operating on open marketplaces. This combination of increased operational efficiency for ransomware groups and reduced risk for IABs has fueled the rapid expansion and influence of IABs within the cyber crime ecosystem. Where are IABs focusing? In 2023, the business services sector was clearly the most targeted industry, although it is still in the top 3 in 2024 with 13% there is a much wider spread of industries being targeted. Whereas in 2023 the business services sector took up a whopping 29% of attacks, that number stood at just 13% in 2024. The same is true for the other industries showing diminished percentages. This could be due to IABs broadening the industries that they are targeting. As usual the USA is a prime target, for its economic and technological power making high value targets. Notably, Brazil and France secured the second and third spots respectively, indicating high value targets in both countries. To see what types of businesses are being targeted in more depth read our guide to IABs here. The Financial Motives of IABs The Initial Access Broker (IAB) market demonstrates a dynamic pricing structure, generally offering corporate access between $500 and $3,000. While 2023 saw an average listing price of $1,979, skewed by occasional high-value targets reaching tens of thousands of dollars, the median price remained significantly lower at $1,000, with a majority of listings below $3,000. In 2024, cybercriminals are increasingly targeting smaller victims. While they've generally lowered the prices for selling access to hacked systems, with 86% costing under $3,000, the average price has actually gone up to $2,047. This higher average is misleading because a few very expensive sales are skewing the number. As the chart shows, the vast majority (58%) of access deals now cost less than $1,000 – a big change from 2023. Furthermore, expensive access options are less common, now making up only 7% of what's for sale. This strategic price reduction, coupled with a decrease in high-value listings, suggests a change in IAB tactics. They are now focusing on volume, offering numerous lower-priced access points that, in aggregate, can yield substantial financial gains. Despite the lower individual prices, the sheer quantity of available access points poses a significant threat, potentially causing widespread damage and proving more lucrative than a smaller number of high-priced sales. This shift indicates an evolution in the IAB market, prioritizing accessibility and volume over individual high-value transactions. To see detailed information on the TTPs being used by IABs, read our guide here. What's next for IABs? The rise of Initial Access Brokers (IABs) is driven by a confluence of factors that enhance the efficiency and profitability of cyber crime. Their specialization in initial network infiltration allows ransomware groups and other malicious actors to focus on later stages of attacks, streamlining operations and increasing the scale of potential damage. The growing trend of direct collaboration between IABs and Ransomware-as-a-Service (RaaS) affiliates further accelerates attack timelines, creating a more efficient and dangerous cyber criminal ecosystem. The evolution of IAB pricing strategies also reveals a significant shift in tactics. IABs are increasingly focusing on volume, offering numerous lower-priced access. This strategy maximizes potential financial gains by providing a wider range of attack vectors, making cyber crime more accessible and potentially more damaging. This shift, coupled with the reduced visibility afforded by operating outside of public dark web forums, provides IABs with a significant layer of protection from law enforcement. Looking ahead, we can expect IABs to continue playing a pivotal role in the cyber crime landscape. Their ability to provide readily available access points will likely fuel the growth of ransomware and other financially motivated attacks. The trend towards lower-priced, high-volume access sales suggests that smaller organizations, previously considered less attractive targets, will face increasing risk. Furthermore as IABs mature their tactics, and strengthen ties with RaaS affiliates, the speed and efficiency of cyber attacks will continue to increase. Therefore, proactive cyber security measures, including threat intelligence on up to date TTPs, continuous monitoring, and employee training, will become increasingly critical in mitigating the growing threat posed by IABs. For detailed insights into contemporary IAB tactics, including access types, privilege usage, and recommended protective measures, consult the comprehensive IAB guide or attend our talk at this year's RSA conference by Adi Bleih, Security Researcher titled Initial Access Brokers – A Deep Dive on April 30th at 2:25pm in HT-W09. You can add it to your schedule here. Found this article interesting? This article is a contributed piece from one of our valued partners. Follow us on Twitter  and LinkedIn to read more exclusive content we post. SHARE    

John Edwards, Technology Journalist & AuthorApril 11, 20255 Min ReadMikhail Reshetnikov via Alamy Stock PhotoEvery IT team seems to have one -- the member who's highly dedicated and talented, yet also something of a free spirit. Knowing how to tolerate and cater to this individual's unique needs without alienating other team members isn't a task generally covered in Management 101 courses for CIOs and IT leaders, yet it's essential in order to keep your team happy and productive. Instead of trying to fit a quirky team member into a rigid mold, work to understand what makes them tick and leverage that unique perspective, suggests Anbang Xu, founder of JoggAI, an AI-powered video platform, and a former senior product manager at Apple and senior software engineer at Google. It’s important to give these individuals space to thrive in their own way, while maintaining clear communication and setting expectations, he observes in an email interview. "By focusing on their strengths, I’ve found that they can bring innovative solutions and fresh ideas that would otherwise be overlooked." Embracing Uniqueness Embrace uniqueness while setting clear expectations, recommends Chetan Honnenahalli, engineering lead at software firm Hubspot and a former team leader at Meta, Zoom, and American Express. "Focus on their strengths and the value they bring to the team but establish boundaries to ensure their behavior doesn’t disrupt team dynamics or project goals," he says in an online interview. "Frequent one-on-one check-ins can help address potential concerns while reinforcing their contributions." Related:Balance respect for individuality with the needs of the team and organization. By valuing their quirks as part of their creative process, you'll foster a sense of belonging and loyalty, Honnenahalli says. "Clear boundaries and open communication will prevent potential misunderstandings, ensuring harmony within the team." Tolerance should depend on the impact of their behavior on team dynamics and project outcomes, Honnenahalli says. "Quirks that enhance creativity or problem-solving should be celebrated, but behaviors that cause disruptions, undermine morale, or create inefficiencies should be addressed promptly." Toleration Techniques Quirky behavior can become an issue if it interferes with the employee's ability to perform their work or if it disrupts fellow team members, says Matt Erhard, managing partner with professional search firm Summit Search Group, via email. "In these cases, the best approach is to have a one-on-one conversation with that employee," he advises. "Address the specific behaviors of concern and establish some expectations and boundaries about what is and isn't acceptable within the workplace." Related:Give the quirky team member strategies and guidelines to adapt their behavior within the workplace setting, Erhard recommends. "It should be made clear that you aren't criticizing or trying to change their personality but rather establishing rules about how they're expected to interact with their colleagues or customers when they're at work." As long as a maverick's behavior doesn't impede team collaboration, project deadlines, or morale, there’s room for individuality, Xu says. "The level of quirkiness you’re willing to tolerate is really a matter of balance," he states. "If their personality adds value without disrupting the team's harmony or performance, then it’s worth embracing." Team Impact Set team norms that allow for individuality while ensuring mutual respect and collaboration, Honnenahalli recommends. Address issues directly and constructively, ensuring open dialogue and fair resolutions. "Highlight how the individual’s quirks contribute positively to the team’s success, encouraging a culture of acceptance." Open communication is vital, Erhard says. "Talk to other team members about the issues they're having and why it's a concern for them." Facilitating a dialogue between the individuals can help both parties see each other’s perspectives. Related:When to Clamp Down Leaders should aim to channel quirkiness constructively rather than working to eliminate it. For instance, if a quirky habit is distracting or counterproductive, the team leader can guide the individual toward alternatives that achieve similar results without causing friction, Honnenahalli says. Avoid suppressing individuality unless it directly conflicts with professional responsibilities or team cohesion. Help the unconventional team member channel their quirks productively rather than trying to reduce them, Xu suggests. "This means offering support and guidance in ways that allow them to thrive within the structure of the team." Remember that quirks can often be a unique asset in problem-solving and innovation. Diverse Perspectives In IT, where innovation thrives on diverse perspectives, quirky team members often deliver creative solutions and unconventional thinking, Honnenahalli says. "Leaders who manage such individuals effectively can cultivate a culture of innovation and inclusivity, boosting morale and productivity." Every team needs a mix of personalities to excel, Xu observes. "The most innovative teams I’ve worked with had a variety of thinkers -- some more conventional, others quirky in their approach." It's the diversity in thinking that drives creativity and breakthroughs. "As leaders, it’s our responsibility to cultivate an environment where these differences are not only accepted but celebrated."About the AuthorJohn EdwardsTechnology Journalist & AuthorJohn Edwards is a veteran business technology journalist. His work has appeared in The New York Times, The Washington Post, and numerous business and technology publications, including Computerworld, CFO Magazine, IBM Data Management Magazine, RFID Journal, and Electronic Design. He has also written columns for The Economist's Business Intelligence Unit and PricewaterhouseCoopers' Communications Direct. John has authored several books on business technology topics. His work began appearing online as early as 1983. Throughout the 1980s and 90s, he wrote daily news and feature articles for both the CompuServe and Prodigy online services. His "Behind the Screens" commentaries made him the world's first known professional blogger.See more from John EdwardsWebinarsMore WebinarsReportsMore ReportsNever Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.SIGN-UPYou May Also Like

About FilestageFilestage frees people from chaotic approval processes, making work more joyful and productive. From large enterprises to independent agencies, our review and approval platform helps teams share, discuss, and approve all their files, all in one place – including documents, designs, images, videos, and audio files.We're a fully remote team with people working in home offices, co-working spaces, and coffee shops all over the world. Together, we're on a mission to create a seamless approval process that helps people deliver their best work. We've raised our Series A and have over half a million users across 500+ companies, including AB InBev, LG, Havas, GroupM, and Emirates. So if you're looking for a fast-growing startup in a booming market, you've found it!What you’ll be working on as our SDR (North America)We’re looking for a Sales Development Representative to help expand our business in the American Market.This role is fully remote and can be based anywhere in the Americas working within EST.In your first year at Filestage, you will:🔍 Spot the right people – You’ll dig into accounts to find the perfect contacts that fit our ideal customer profile (ICP) and bring the highest potential value.📞 Pick up the phone – Cold calling is your superpower. With around 100 calls a day, you’ll be actively creating new opportunities and opening doors.💌 Go multi-channel – You won’t just stick to the phone. You’ll also use smart email sequences and social selling to connect with prospects in meaningful ways.💡 Spark interest – Through great conversations, you’ll uncover pain points and show prospects how we can help—setting up high-quality demo meetings for the sales team.🔄 Keep the flow smooth – You’ll ensure seamless handovers to AEs by clearly documenting your outreach and helping to keep the pipeline healthy and predictable.🧪 Experiment & improve – You’ll constantly test and tweak your outreach scripts and sequences to find what works best—and then double down.📈 Share what you learn – You’ll pass on insights to the Head of Sales, helping to refine our sales process and boost conversion rates across the board.🎯 Own your results – You’ll track your own KPIs, spot trends early (good or bad), and keep your manager in the loop so we can always stay one step ahead.Life at FilestageWe believe people are more productive when they can choose their own schedule. So we’re proud to offer fully-remote roles that give you the perfect balance between work and life.Here are some of the benefits you can look forward to at Filestage:Work from where you’re happiest and enjoy a flexible schedule. We’ve been fully remote from the start, giving you the opportunity to meet people all over the world and broaden your horizons.For this role, we’re looking for someone based in EST time zone to make sure there’s a high time overlap with our team members, customers, and partners.Meet up in real life. We all travel together at least once a year (in normal times) at our team retreat to have fun and get to know each other.Enjoy a strong team culture. We’re a group of knowledge seekers, reflective thinkers, clear communicators, goal owners, problem solvers, and team players. These are the values we strive for to help us achieve our mission.Join a happy team. We’ve been rated five stars on Glassdoor by our lovely team. You can take a look at our reviews here.Create a workspace that suits you. You’ll get a €1,500 budget for hardware, as well €500 for home office to buy whatever you need to do your best work – including a computer, webcam, or standing desk.Get 38 days of holiday. Plenty of time for city breaks, summer escapes, and everything in between. You’ll also get a half day on your birthday to give you a chance to celebrate!Volunteer/Charity Day.  Enjoy a Volunteer/Charity Day to support a cause close to your heart.Continue to grow and develop your career. We care about your development and want you to be able to learn new things! After six months in the company, you’ll get a budget to be able to use for personal development.Make your voice heard. We trust our team members to make the best decisions to achieve their goals, so you won’t have to put up with micromanagers here.Say goodbye to pointless meetings. We practice what we preach when it comes to productivity, so you can expect flat hierarchies, fast iterations, and no bullshit meetings.

For much of last year, about 2,500 US service members from the 15th Marine Expeditionary Unit sailed aboard three ships throughout the Pacific, conducting training exercises in the waters off South Korea, the Philippines, India, and Indonesia. At the same time, onboard the ships, an experiment was unfolding: The Marines in the unit responsible for sorting through foreign intelligence and making their superiors aware of possible local threats were for the first time using generative AI to do it, testing a leading AI tool the Pentagon has been funding. Two officers tell us that they used the new system to help scour thousands of pieces of open-source intelligence—nonclassified articles, reports, images, videos—collected in the various countries where they operated, and that it did so far faster than was possible with the old method of analyzing them manually. Captain Kristin Enzenauer, for instance, says she used large language models to translate and summarize foreign news sources, while Captain Will Lowdon used AI to help write the daily and weekly intelligence reports he provided to his commanders.  “We still need to validate the sources,” says Lowdon. But the unit’s commanders encouraged the use of large language models, he says, “because they provide a lot more efficiency during a dynamic situation.” The generative AI tools they used were built by the defense-tech company Vannevar Labs, which in November was granted a production contract worth up to $99 million by the Pentagon’s startup-oriented Defense Innovation Unit with the goal of bringing its intelligence tech to more military units. The company, founded in 2019 by veterans of the CIA and US intelligence community, joins the likes of Palantir, Anduril, and Scale AI as a major beneficiary of the US military’s embrace of artificial intelligence—not only for physical technologies like drones and autonomous vehicles but also for software that is revolutionizing how the Pentagon collects, manages, and interprets data for warfare and surveillance.  Though the US military has been developing computer vision models and similar AI tools, like those used in Project Maven, since 2017, the use of generative AI—tools that can engage in human-like conversation like those built by Vannevar Labs—represent a newer frontier. The company applies existing large language models, including some from OpenAI and Microsoft, and some bespoke ones of its own to troves of open-source intelligence the company has been collecting since 2021. The scale at which this data is collected is hard to comprehend (and a large part of what sets Vannevar’s products apart): terabytes of data in 80 different languages are hoovered every day in 180 countries. The company says it is able to analyze social media profiles and breach firewalls in countries like China to get hard-to-access information; it also uses nonclassified data that is difficult to get online (gathered by human operatives on the ground), as well as reports from physical sensors that covertly monitor radio waves to detect illegal shipping activities.  Vannevar then builds AI models to translate information, detect threats, and analyze political sentiment, with the results delivered through a chatbot interface that’s not unlike ChatGPT. The aim is to provide customers with critical information on topics as varied as international fentanyl supply chains and China’s efforts to secure rare earth minerals in the Philippines.  “Our real focus as a company,” says Scott Philips, Vannevar Labs’ chief technology officer, is to “collect data, make sense of that data, and help the US make good decisions.”  That approach is particularly appealing to the US intelligence apparatus because for years the world has been awash in more data than human analysts can possibly interpret—a problem that contributed to the 2003 founding of Palantir, a company now worth nearly $217 billion and known for its powerful and controversial tools, including a database that helps Immigration and Customs Enforcement search for and track information on undocumented immigrants.  In 2019, Vannevar saw an opportunity to use large language models, which were then new on the scene, as a novel solution to the data conundrum. The technology could enable AI not just to collect data but to actually talk through an analysis with someone interactively. Vannevar’s tools proved useful for the deployment in the Pacific, and Enzenauer and Lowdon say that while they were instructed to always double-check the AI’s work, they didn't find inaccuracies to be a significant issue. Enzenauer regularly used the tool to track any foreign news reports in which the unit’s exercises were mentioned and to perform sentiment analysis, detecting the emotions and opinions expressed in text. Judging whether a foreign news article reflects a threatening or friendly opinion toward the unit is a task that on previous deployments she had to do manually. “It was mostly by hand—researching, translating, coding, and analyzing the data,” she says. “It was definitely way more time-consuming than it was when using the AI.”  Still, Enzenauer and Lowdon say there were hiccups, some of which would affect most digital tools: The ships had spotty internet connections much of the time, limiting how quickly the AI model could synthesize foreign intelligence, especially if it involved photos or video.  With this first test completed, the unit’s commanding officer, Colonel Sean Dynan, said on a call with reporters in February that heavier use of generative AI was coming; this experiment was “the tip of the iceberg.”  This is indeed the direction that the entire US military is barreling toward at full speed. In December, the Pentagon said it will spend $100 million in the next two years on pilots specifically for generative AI applications. In addition to Vannevar, it’s also turning to Microsoft and Palantir, which are working together on AI models that would make use of classified data. (The US is of course not alone in this approach; notably, Israel has been using AI to sort through information and even generate lists of targets in its war in Gaza, a practice that has been widely criticized.) Perhaps unsurprisingly, plenty of people outside the Pentagon are warning about the potential risks of this plan, including Heidy Khlaaf, who is chief AI scientist at the AI Now Institute, a research organization, and has expertise in leading safety audits for AI-powered systems. She says this rush to incorporate generative AI into military decision-making ignores more foundational flaws of the technology: “We’re already aware of how LLMs are highly inaccurate, especially in the context of safety-critical applications that require precision.”  One particular use case that concerns her is sentiment analysis, which she argues is “a highly subjective metric that even humans would struggle to appropriately assess based on media alone.”  If AI perceives hostility toward US forces where a human analyst would not—or if the system misses hostility that is really there—the military could make an misinformed decision or escalate a situation unnecessarily. Sentiment analysis is indeed a task that AI has not perfected. Philips, the Vannevar CTO, says the company has built models specifically to judge whether an article is pro-US or not, but MIT Technology Review was not able to evaluate them.  Chris Mouton, a senior engineer for RAND, recently tested how well-suited generative AI is for the task. He evaluated leading models, including OpenAI’s GPT-4 and an older version of GPT fine-tuned to do such intelligence work, on how accurately they flagged foreign content as propaganda compared with human experts. “It’s hard,” he says, noting that AI struggled to identify more subtle types of propaganda. But he adds that the models could still be useful in lots of other analysis tasks.  Another limitation of Vannevar’s approach, Khlaaf says, is that the usefulness of open-source intelligence is debatable. Mouton says that open-source data can be “pretty extraordinary,” but Khlaaf points out that unlike classified intel gathered through reconnaissance or wiretaps, it is exposed to the open internet—making it far more susceptible to misinformation campaigns, bot networks, and deliberate manipulation, as the US Army has warned. For Mouton, the biggest open question now is whether these generative AI technologies will be simply one investigatory tool among many that analysts use—or whether they’ll produce the subjective analysis that’s relied upon and trusted in decision-making. “This is the central debate,” he says.  What everyone agrees is that AI models are accessible—you can just ask them a question about complex pieces of intelligence, and they’ll respond in plain language. But it’s still in dispute what imperfections will be acceptable in the name of efficiency. 

The Royal Borough of Kensington and Chelsea (RBKC) has launched a series of lawsuits against firms involved in the refurbishment of Grenfell Tower. According to a report in the Financial Times, the local authority is pursuing £360m plus interest from suppliers, subcontractors and other companies. Seven companies are being sued, according to the report, including Arconic Architectural Products, Artelia Projects UK, Celotex, CEP Architectural Facades, Harley Facades, Rydon Maintenance, and Whirlpool (now known as Beko). Source: ShutterstockThe Grenfell fire killed 72 people in June 2017 In court documents cited by the newspaper, RBKC specified losses including property acquisition and renovation costs of £129m, housing and temporary accommodation expenses of £47m and employment and staff costs of £41m. Other costs referenced included £18m worth of hardship payments and £970,000 for fire marshals. A Kensington and Chelsea council spokesperson said: “We have issued legal proceedings against a number of companies, in line with the council’s ongoing commitment to ensure those parties pay a share of the costs incurred against the public purse.” The council’s lawsuits come after the publication of the Grenfell Inquiry phase 2 report last September. > Also read: MHCLG must take ‘clearer central role’ to fix remediation, says Clive Betts > Also read: How closely has the government followed the Grenfell Inquiry’s recommendations? The report blamed the fire, which resulted in the deaths of 72 people in June 2017, on decades of government failure, a slapdash construction culture and the dishonesty of building product manufacturers. The inquiry found that the choice of combustible materials for the cladding on the tower “resulted from a series of errors caused by the incompetence of the organisations and individuals involved in the refurbishment”. It also found that the borough had neglected resident concerns about the tower and that its building control department had failed to ensure compliance. All the firms involved in the suit have been contacted for comment.