Nadella had been Microsoft CEO for less than two years. Azure, the company's cloud platform, was five years old and chasing Amazon Web Services for market share. And now AWS had been listed as a donor in the “Introducing OpenAI” post. In the internal message, which hasn't been previously reported, Nadella wondered how the new AI nonprofit could remain truly “open” if it was tied only to Amazon's cloud. New court filings offer an inside look at one of the most consequential relationships in tech. For this story, GeekWire dug through more than 200 documents, many of them made public Friday in Elon Musk's ongoing suit accusing OpenAI and its CEO Sam Altman of abandoning the nonprofit mission. Musk, who was an OpenAI co-founder, is seeking up to $134 billion in damages. OpenAI has disputed Musk's account of the company's origins. In a blog post last week, the company said Musk agreed in 2017 that a for-profit structure was necessary, and that negotiations ended only when OpenAI refused to give him full control. In March 2018, after learning of OpenAI's plans to launch a commercial arm, Microsoft CTO Kevin Scott sent Nadella and others an email offering his thoughts. “I wonder if the big OpenAI donors are aware of these plans?” Scott wrote. “Ideologically, I can't imagine that they funded an open effort to concentrate ML [machine learning] talent so that they could then go build a closed, for profit thing on its back.” The latest round of documents, filed as exhibits in Musk's lawsuit, represents a partial record selected to support his claims in the case. Microsoft's relationship with OpenAI has been one of its key strategic advantages in the cloud. But the behind-the-scenes emails make it clear that Amazon was actually there first. According to an internal Microsoft slide deck from August 2016, included in recent filings, OpenAI was running its research on AWS as part of a deal that gave it $50 million in computing for $10 million in committed funds. In late August, the filings show, Altman emailed Musk about a new deal with Microsoft: “I have negotiated a $50 million compute donation from them over the next 3 years!” he wrote. “Do you have any reason not to like them, or care about us switching over from Amazon?” Musk, co-chair of OpenAI at the time, gave his blessing to the Microsoft deal in his unique way, starting with a swipe at Amazon founder Jeff Bezos: “I think Jeff is a bit of a tool and Satya is not, so I slightly prefer Microsoft, but I hate their marketing dept,” Musk wrote. … And their offering wasn't that good technically anyway.” Microsoft and OpenAI announced their partnership in November 2016 with a blog post highlighting their plans to “democratize artificial intelligence,” and noting that OpenAI would use Azure as its primary cloud platform going forward. The August 2016 slide deck, titled “OpenAI on Azure Big Compute,” described it as a prime opportunity to flip a high-profile customer to Azure. The presentation also emphasized bigger goals: “thought leadership” in AI, a “halo effect” for Azure's GPU launch, and the chance to recruit a “net-new audience” of developers and startups. It noted that OpenAI was a nonprofit “unconstrained by a need to generate financial return” — an organization whose research could burnish Microsoft's reputation in AI. In June 2017, Musk spoke with Nadella directly to pitch a major expansion. Nadella said he'd talk about it internally with his Microsoft cloud team, according to the email. “Sounds like there is a good chance they will do it,” Musk wrote. Two months later, Altman followed up with a formal pitch. “I think it will be the most impressive thing yet in the history of AI,” he wrote to Nadella that August. Microsoft's cloud executives ran the numbers and balked. “I do believe the pop from someone like Sam and Elon will help build momentum for Azure,” Zander wrote. But I won't take a complete bath to do it.” OpenAI contracted with Google for the Dota 2 project instead. Microsoft's broader relationship with OpenAI was starting to fray, as well. By January 2018, according to internal emails, Microsoft executive Brett Tanzer had told Altman that he was having a hard time finding internal sponsors at Microsoft for an expanded OpenAI deal. Around that time, Tanzer noted in an email to Nadella and other senior executives that OpenAI's people “have been up in the area recently across the lake” — a reference to Amazon's Seattle headquarters. Scott wrote that OpenAI was treating Microsoft “like a bucket of undifferentiated GPUs, which isn't interesting for us at all.” Harry Shum, who led Microsoft's AI research, said he'd visited OpenAI a year earlier and “was not able to see any immediate breakthrough in AGI.” Eric Horvitz, Microsoft's chief scientist, chimed in to say he had tried a different approach. After a Skype call with OpenAI co-founder Greg Brockman, he pitched the idea of collaboration focused on “extending human intellect with AI — versus beating humans.” The conversation was friendly, Horvitz wrote, but he didn't sense much interest. He suspected OpenAI's Dota work was “motivated by a need to show how AI can crush humans, as part of Elon Musk's interest in demonstrating why we should all be concerned about the power of AI.” Scott summed up the risk of walking away: OpenAI might “storm off to Amazon in a huff and shit-talk us and Azure on the way out.” “All things equal, I'd love to have them be a Microsoft and Azure net promoter. Not sure that alone is worth what they're asking.” But by the following year, Microsoft had found a reason to double down. The nonprofit would remain, but a new “capped profit” entity would sit beneath it — a hybrid that could raise capital from investors while limiting their returns. Microsoft agreed to invest $1 billion, with an option for a second billion, in exchange for exclusive cloud computing rights and a commercial license to OpenAI's technology. In a June 2019 email to Nadella and Bill Gates, previously disclosed in the Google antitrust case, Scott cited the search giant's AI progress as one reason for Microsoft to invest in OpenAI. Nadella forwarded Scott's email to Amy Hood, Microsoft's CFO. “Very good email that explains why I want us to do this,” Nadella wrote, referring to the larger OpenAI investment, “and also why we will then ensure our infra folks execute.” According to Nadella's deposition testimony, the Microsoft co-founder was clear in “wanting us to just do our own” — arguing that the company should focus on building AI capabilities in-house rather than placing such a large bet on OpenAI. Nadella explained that the decision to invest was eventually driven by him and Scott, who concluded that OpenAI's specific research direction into transformers and large language models (the GPT class) was more promising than other approaches at the time. Hood, meanwhile, offered some blunt commentary on OpenAI's cap on profits — the centerpiece of its new structure, meant to limit investor returns and preserve the nonprofit's mission. “Given the cap is actually larger than 90% of public companies, I am not sure it is terribly constraining nor terribly altruistic but that is Sam's call on his cap,” Hood wrote in a July 14, 2019, email to Nadella, Scott, and other executives. If OpenAI succeeded, she noted, the real money for Microsoft would come from Azure revenue — far exceeding any capped return on the investment itself. But the deal gave Microsoft more than cloud revenue. Under the partnership agreement, major decisions required approval from a majority of limited partners based on how much they had contributed. In September 2020, Musk responded to reports that Microsoft had exclusively licensed OpenAI's GPT-3. “This does seem like the opposite of open,” he tweeted. In an October 2020 meeting, according to internal notes cited in a recent court order, Microsoft executives discussed the perception that the company was “effectively owning” OpenAI, with Nadella saying they needed to give thought to Musk's perspective. In February 2021, as Microsoft and OpenAI negotiated a new investment, Altman emailed Microsoft's team: “We want to do everything we can to make you all commercially successful and are happy to move significantly from the term sheet.” His preference, Altman told the Microsoft execs, was “to make you all a bunch of money as quickly as we can and for you to be enthusiastic about making this additional investment soon.” They closed the deal in March 2021, for up to $2 billion. This was not disclosed publicly until January 2023, when Microsoft revealed it as part of a larger investment announcement. According to a transcript of her deposition, Mira Murati, then OpenAI's vice president of applied AI and partnerships, had written in contemporaneous notes that the most-cited goal inside the company that year was a $100 million revenue target. Altman had told employees that Nadella and Scott said this needed to be hit to justify the next investment, as much as $10 billion. Murati testified that Altman told her “it was important to achieve this goal to receive Microsoft's continued investments.” OpenAI responded by expanding its go-to-market team and building out its enterprise business. It was the moment that turned OpenAI from an AI research lab into a household name. OpenAI's board learned about the launch on Twitter. According to deposition testimony, board members Helen Toner and Tasha McCauley received no advance notice and discovered ChatGPT by seeing screenshots on social media. Altman, according to one filing, characterized the release as a “research preview” using existing technology. As its biggest investor, Microsoft pushed OpenAI to monetize the product's success. In mid-January 2023, Nadella texted Altman asking when they planned to activate a paid subscription. the only real reason for rushing it is we are just so out of capacity and delivering a bad user experience.” He asked Nadella for his input: “any preference on when we do it?” Two weeks later, Nadella checked in again: “Btw …how many subs have you guys added to chatGPT?” OpenAI had 6 million daily active users — their capacity limit — and had turned away 50 million people who tried to sign up. The product's viral success validated Microsoft's bet and foreshadowed a new era of demand for its cloud platform. Murati, then OpenAI's chief technology officer and the board's choice for interim CEO, called Nadella and Kevin Scott to warn them just 10 to 15 minutes before Altman himself was told. “Mira sounded like she had been run over by a truck as she tells me,” Scott wrote in an email to colleagues that weekend. They had given her less than 24 hours to prepare. An all-hands meeting followed at 2 p.m. By Friday night, Brockman had resigned. A “whole horde” of employees, Scott wrote, had reached out to Altman and Brockman “expressing loyalty to them, and saying they will resign.” Microsoft didn't have a seat on the board. But text messages between Nadella and Altman, revealed in the latest filings, show just how influential it was in the ultimate outcome. At 7:42 a.m. Pacific on Saturday, Nov. 18, Nadella texted Altman asking if he was free to talk. Altman replied that he was on a board call. That evening, at 8:25 p.m., Nadella followed up with a detailed message from Brad Smith, Microsoft's president and top lawyer. In a matter of hours, the trillion-dollar corporation had turned on a dime, establishing a new subsidiary from scratch — legal work done, papers ready to file as soon as the Washington Secretary of State opened Monday morning. They called it Microsoft RAI Inc., using the acronym for Responsible Artificial Intelligence. “We can then capitalize the subsidiary and take all the other steps needed to operationalize this and support Sam in whatever way is needed,” Smith wrote. By Sunday, Emmett Shear, the Twitch co-founder, had replaced Murati as interim CEO. That night, when the board still hadn't reinstated Altman, Nadella announced publicly that Microsoft was prepared to hire the OpenAI CEO and key members of his team. “In a world of bad choices,” Nadella said in his deposition, the move “was definitely not my preferred thing.” But it was preferable to the alternative, he added. “The worst outcome would have been all these people leave and they go to our competition.” Altman messaged Nadella and Scott that morning, “can we talk soon? The other slots were “still up in air.” Altman asked, “would [redacted] be ok with you?” One candidate, the Microsoft president wrote, was “Solid, thoughtful, calm.” Another was “Incredibly smart, firm, practical, while also a good listener.” At one point, Scott floated a joke: “I can quit for six months and do it.” He added a grinning emoji and commented, “Ready to be downvoted by Satya on this one, and not really serious.” Nadella had said publicly that the company didn't want one. “Your future would be decided by Larry [Summers],” he wrote. “He's smart but so mercurial.” He called it “too risky.” (Summers resigned from the OpenAI board in November 2025, following revelations about his correspondence with Jeffrey Epstein.) Later that night, OpenAI announced Altman's return with the newly constituted board. On December 27, 2024, OpenAI announced it would unwind its capped-profit structure. Internally, this initiative was called “Project Watershed,” the documents reveal. Its approval rights over “Major Decisions” including changes to OpenAI's structure. A SoftBank-led financing — internally called Project Sakura — was contingent on the recapitalization closing by year-end. Goldman Sachs, advising OpenAI, put it at $353 billion. As Altman put it in his deposition, “That was the willing buyer-willing seller market price, so I won't argue with it.” For Microsoft, it was a give-and-take deal: the tech giant lost its right of first refusal on new cloud workloads, even as OpenAI committed to the $250 billion in future Azure purchases. Now any such declaration needs to be made by an independent panel, and Microsoft's IP rights run through 2032 regardless. Six days later, OpenAI signed a seven-year, $38 billion infrastructure deal with Amazon Web Services. The company that had “sneaked in there” at the founding, as Nadella put it in 2015, was back — this time as a major cloud provider for Microsoft's flagship AI partner. In a post this weekend, OpenAI CFO Sarah Friar made the shift explicit: “Three years ago, we relied on a single compute provider,” she wrote. “Today, we are working with providers across a diversified ecosystem. That shift gives us resilience and, critically, compute certainty.” OpenAI is no longer a research lab dependent on Microsoft's cloud. A decade later, nothing could happen without the Redmond tech giant. But OpenAI will no longer be theirs alone. GeekWire Studios has partnered with AWS for the Guide to re:Invent. This interview series took place on the Expo floor at AWS re:Invent 2025, and features insightful conversations about the future of cloud tech, as well as partnership success stories. Click for more about underwritten and sponsored content on GeekWire. Have a scoop that you'd like GeekWire to cover? Tensions with OpenAI underscore Microsoft's need to ensure its own AI future Microsoft and OpenAI make a deal: Reading between the lines of their secretive new agreement Elon and Satya, together again: Microsoft brings Musk's xAI models to Azure, despite OpenAI feud

“AI is already the most consequential technological shift of my lifetime,” she wrote on LinkedIn. “It will shape our children's lives in ways that are difficult to predict.” Before joining Pinterest in 2018, Mallard was CMO at Athleta and Omada Health. Cheever spent nearly a decade at Amazon in global marketing roles, where she oversaw various functions and campaigns. “What really hooked me is how Pinterest stands apart from other social and search platforms,” Cheever wrote on LinkedIn. Jo Shoesmith, global chief creative officer at Amazon, will replace Cheever on an interim basis as the company searches for a permanent replacement. “Claudine has been a creative leader, building durable global brand architecture and sophisticated creative systems that operate at scale across our Stores business,” an Amazon spokesperson said in a statement. San Francisco-based Pinterest also hired former DoorDash and Spotify exec Lee Brown as its new chief business officer. — Irina Ghose was named managing director for Anthropic's India operations. India represents the second-largest user base for Anthropic's Claude product, Ghose wrote on LinkedIn. — Trisha Berard, a longtime Seattle-area HR leader, joined Evergreen Goodwill as senior vice president of people and culture. Evergreen Goodwill employs more than 2,000 people across the Seattle region and operates 23 nonprofit thrift stores. Libby Johnson McKee, a former customer returns leader at Amazon, joined Evergreen Goodwill as CEO in 2024. Shelley Salomon, vice president of global business for Amazon, sits on the nonprofit's board. The chips powering your smart TV, voice assistant, tablet, and car all have something in common: MediaTek Click for more about underwritten and sponsored content on GeekWire. Google makes a big move into agentic commerce, raising questions about Amazon's retail dominance GeekWire Podcast: Alexa's next act, Microsoft's retail play, Google's AI Inbox, and a smart bird feeder fail GeekWire Studios has partnered with AWS for the Guide to re:Invent. This interview series took place on the Expo floor at AWS re:Invent 2025, and features insightful conversations about the future of cloud tech, as well as partnership success stories. Click for more about underwritten and sponsored content on GeekWire. Seattle tech vets Marco Matos and Kabir Shahani rethink marketing for big brands with new AI startup Tech Moves: Longtime Google leader Umesh Shankar joins Microsoft as new corporate VP Tech Moves: Expedia names first AI chief; Textio founder joins Microsoft; T-Mobile exec departs

TPC is taking over 6,600 square feet of space directly above its current location in the historic Guiry Schillestad Building at 2101 1st Ave. “Even with the challenges downtown has faced, Belltown has been a great neighborhood for us,” TPC CEO Christopher Hoyt said in a news release. The expansion will give TPC extra meeting room capacity and additional private offices for teams of up to 10 employees. GeekWire Studios has partnered with AWS for the Guide to re:Invent. This interview series took place on the Expo floor at AWS re:Invent 2025, and features insightful conversations about the future of cloud tech, as well as partnership success stories. Click for more about underwritten and sponsored content on GeekWire. Pioneer Collective to open new co-working space at development along Ship Canal in Seattle Co-working company Kiln opening first Washington state location in tech-heavy Seattle neighborhood Amazon HQ building will become new home for longtime Seattle ceramics nonprofit Inside Seattle's newest co-working space, where Centrl Office is taking over former WeWork location

As farmers grapple with extreme weather, supply chain disruptions and labor shortages, Washington state is betting that artificial intelligence could help secure the future of agriculture. “So what if Washington can get ahead of this and be intentional about how we use AI in agriculture?” There are already a number of AI-driven, ag tech companies based in Washington, including Carbon Robotics, which manufactures autonomous farming machines that zap weeds with lasers. “Farmers and technologists see the world in similar ways,” Mikesell said. We tackle problems head on, put in a lot of hard work …. To be successful in this space, he emphasized the importance of genuinely partnering with farmers to learn their specific challenges rather than coming in with predetermined solutions. This is the first time the Academy of Sciences, which educates public leaders on scientific matters, has created a series focused on one issue and incorporated a call to action. Farming hasn't been made a priority compared to other areas, he added, but the need is urgent and Washington can help lead. Microsoft's mission: empowering every person and organization on the planet to achieve more. Learn how Microsoft is thinking about responsible artificial intelligence, regulation, sustainability, and fundamental rights. AI research boost: University of Washington expands infrastructure with $10M in federal funding What Microsoft has learned about housing, and why it's urging the state to unlock commercial land GeekWire Studios has partnered with AWS for the Guide to re:Invent. This interview series took place on the Expo floor at AWS re:Invent 2025, and features insightful conversations about the future of cloud tech, as well as partnership success stories. Carbon Robotics raises $70M for AI robots that blast weeds with lasers Orchard Robotics, an agtech startup using AI to solve farming's data problem, raises $22M Carbon Robotics unveils ‘AutoTractor,' a self-driving platform to boost productivity on farms Carbon Robotics raises $20M as LaserWeeder maker plans secretive new ‘AI robot' for farms

Investors in the round include chipmaker Nvidia, Amazon founder Jeff Bezos, and VC firms SV Angel, Google Ventures, and Laurene Powell Jobs' firm Emerson Collective. The megadeal for the three-month-old company follows a trend of investors throwing money at startups founded by breakaways of major AI labs. Humans&'s founders include Andi Peng, a former Anthropic researcher who worked on reinforcement learning and post-training of Claude 3.5 through 4.5; Georges Harik, Google's seventh employee, who helped build its first advertising systems; Eric Zelikman and Yuchen He, two former xAI researchers who helped develop the Grok chatbot; and Noah Goodman, a Stanford professor of psychology and computer science. Humans&'s 20-odd employees also come from OpenAI, Meta, Reflection, AI2, and MIT, according to the company. The startup aims to use software to help people collaborate with each other — think an AI version of an instant messaging app. In order to build AI that serves “as a deeper connective tissue that strengthens organizations and communities,” Humans& hopes to rethink “how we train models at scale and how people interact with AI,” per the company's web page. Every weekday and Sunday, you can get the best of TechCrunch's coverage. TechCrunch Mobility is your destination for transportation news and insight. Startups are the core of TechCrunch, so get our best coverage delivered weekly. Provides movers and shakers with the info they need to start their day. By submitting your email, you agree to our Terms and Privacy Notice.

SpaceX didn't properly inspect a hydraulic crane that had recently been repaired before it collapsed at the company's Starbase, Texas facility last June, according to the Occupational Safety and Health Administration. As a result, the federal safety agency has hit SpaceX with seven “serious” violations related to its investigation, which it opened one day after the crane collapse. OSHA issued the maximum financial penalty it can levy on six of those “serious” violations, giving Elon Musk's spaceflight company a combined fine of $115,850. OSHA's probe is still open according to the agency's website. It's still unclear whether any workers were harmed in the accident. SpaceX has had a long history of injuries at the South Texas launch complex. A 2023 Reuters report uncovered dozens of previously-unreported injuries — and one employee death — that happened as the facility was built up across the last decade. A TechCrunch analysis of OSHA data last year showed SpaceX has a much higher injury rate at the Starbase site than at other facilities run by the company, or those of its industry peers. OSHA is also investigating that accident, as TechCrunch first reported. That Grove crane had recently been repaired, according to the citations, and returned to SpaceX “without the employer ensuring repairs met manufacturer's criteria via an inspection by a qualified person.” It is not clear why the crane was repaired, but another citation states a Grove RT9150E crane at the test site had a computer that would not start until “multiple attempts” were made. OSHA claims SpaceX was either not performing or not documenting monthly inspections on the Grove crane, and that it hadn't been inspected “within the last 12 months.” SpaceX did not perform monthly inspections of the wire rope being used to move debris either, according to OSHA. And OSHA claims rigging equipment used at the site was missing manufacturer-prescribed markings that are supposed to detail a “safe working load.” As part of its investigation, OSHA also found that an employee had been operating a Tadano 90-ton crawler crane to move debris at the test site with an expired certification from the National Commission for the Certification of Crane Operators. Most recently, he was a reporter at Bloomberg News where he helped break stories about some of the most notorious EV SPAC flops. Why Silicon Valley is really talking about fleeing California (it's not the 5%) AI cloud startup Runpod hits $120M in ARR — and it started with a Reddit post Mira Murati's startup, Thinking Machines Lab, is losing two of its co-founders to OpenAI AI models are starting to crack high-level math problems

In an effort to sweeten the pot for Warner Bros. Netflix said it would finance the deal with cash, debt, and “committed financing.” The change-up comes as rival suitor Paramount Skydance has intensified efforts to win over WBD's shareholders with its all-cash, $30-per-share offer for the entirety of the company, including securing a $40 billion guarantee from its CEO David Ellison's billionaire dad, Oracle co-founder Larry Ellison. Discovery for months, last week sued the company for more information on Netflix's offer and said it would nominate new members to Warner Bros.' board, after WBD rebuffed its offer. Netflix, for its part, had until now stuck to its original cash-and-share offer, enjoying the full backing of WBD's board, which has resolutely rejected Paramount's bids. WBD has argued that selling to Netflix would make for a better deal because the streaming giant has the capital to pay, and has said that Paramount's deal poses “materially more risk,” as it would saddle the combined company with $87 billion in debt. Warner Bros. has also called into question Paramount's ability to function after the deal goes through, arguing that raising such amounts of debt would further worsen Paramount's current “junk” credit rating, and has raised concerns about Paramount's negative free cash flow, which would be exacerbated by the acquisition. WBD said in October that it was exploring a sale after receiving unsolicited interest from multiple parties. The company, valued at over $45 billion at the time but burdened with billions in debt, has struggled amid declining cable viewership and intensifying competition from streaming rivals like Netflix. You can contact or verify outreach from Ram by emailing ram.iyer@techcrunch.com. Why Silicon Valley is really talking about fleeing California (it's not the 5%) AI cloud startup Runpod hits $120M in ARR — and it started with a Reddit post Mira Murati's startup, Thinking Machines Lab, is losing two of its co-founders to OpenAI

The startup was previously valued at $100 million post-money, another source told TechCrunch. Prosus, Lightspeed Venture Partners, Together, and Y Combinator also participated. Emergent has now raised $100 million within seven months of its launch. The startup said it is targeting ARR of more than $100 million by April 2026. “We continue to see massive demand across our top geographies — the U.S., Europe and India — and we'll continue to expand deeper into these markets,” founder Mukund Jha told TechCrunch, adding that the startup's recently launched mobile app-building service is seeing strong adoption. The startup is hiring aggressively across functions in both countries, Jha said. Accel also backed Rocket, another India-founded startup, in a $15 million seed round last year, along with Together Fund and Salesforce Ventures. The deal is also notable as it marks SoftBank's return to investing in India — the firm previously backed Indian commerce startup ElasticRun nearly four years ago. Jagmeet covers startups, tech policy-related updates, and all other major tech-centric developments from India for TechCrunch. You can contact or verify outreach from Jagmeet by emailing mail@journalistjagmeet.com. Why Silicon Valley is really talking about fleeing California (it's not the 5%) AI cloud startup Runpod hits $120M in ARR — and it started with a Reddit post Parloa triples its valuation in 8 months to $3B with $350M raise Mira Murati's startup, Thinking Machines Lab, is losing two of its co-founders to OpenAI

PSMC follows Nanya and Inotera's footsteps, but will retain independence. When you purchase through links on our site, we may earn an affiliate commission. In an unexpected turn of events, Micron announced plans to buy Powerchip Semiconductor Manufacturing Corporation's (PSMC) P5 fabrication site in Tongluo, Miaoli County, Taiwan, for a total cash consideration of $1.8 billion. To a large degree, the transaction would evolve Micron's long-term 'technology-for-capacity' strategy, which it has used for decades. This also signals that DRAM fabs are now so capital-intensive that it is no longer viable for companies like PSMC to build them and get process technologies from companies like Micron. "This strategic acquisition of an existing cleanroom complements our current Taiwan operations and will enable Micron to increase production and better serve our customers in a market where demand continues to outpace supply," said Manish Bhatia, executive vice president of global operations at Micron Technology. "The Tongluo fab's close proximity to Micron's Taichung site will enable synergies across our Taiwan operations." By today's standards, a 300,000 square foot cleanroom is a relatively large one, but it will be dwarfed by Micron's next-generation DRAM campus in New York, which will feature four cleanrooms of 600,000 square feet each. The transaction is expected to close by Q2 2026, pending receipt of all necessary approvals. After closing, Micron will gradually equip and ramp the site for DRAM production, with meaningful wafer output starting in the second half of 2027. The agreement also establishes a long-term strategic partnership under which PSMC will support Micron with assembly services, while Micron will assist PSMC's legacy DRAM portfolio. While the P5 site in Tongluo isn't producing memory in high volumes today, the change of ownership and inevitable upgrade of the fab itself will have an impact on global DRAM supply, which is good news for a segment that is experiencing unprecedented demand. While it is important that Micron is set to buy a production facility in Taiwan, it is even more important that the transaction marks an end to its technology-for-capacity approach to making memory on the island. In the past, instead of building large amounts of new greenfield fabs in Taiwan, Micron partnered with local foundries (most notably PSMC, but also with Inotera and Nanya) and provided advanced DRAM process technology in exchange for wafer capacity, manufacturing services, or fab access. This approach allowed Micron to expand output faster and with less capital risk, leveraged Taiwan's mature 300mm manufacturing ecosystem, and avoided duplicating the front-end infrastructure, which was already in place. It worked well when DRAM fabs cost a few billion dollars, when process ramps were straightforward, and when partners could justify their capital risks in exchange for technologies (which cost billions in R&D investments) and stable wafer demand. Today's advanced DRAM fabs require $15 – $25 billion or more of upfront investment. In that environment, a partner running someone else's IP absorbs massive CapEx and execution risk while getting limited advantages, which makes the economics increasingly unattractive: after all, if you can invest over $20 billion in a fab, you can certainly invest $2 billion in R&D. Early technology-for-capacity deals helped it scale quickly, but once fabs crossed a certain cost and complexity threshold, Micron had to move on and own fabs instead of renting capacity. This is reflected in moves like its Elpida acquisition in 2013, where the company purchased a bankrupt memory maker to secure the company's capacity. What remains of the model is essentially pushed to legacy nodes, which are almost all fully depreciated, or to specialty DRAM, which does not require leading-edge process nodes. Other remnants of the model may be found in additional backend services, where capital intensity and strategic risks are also lower. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher.

Wikipedia's Jimmy Wales has been called the last decent tech baron. It's sounds like a flattering label, although one I usually associate more with yacht-dwelling meatheads who feed their herds of cattle homegrown macadamia nuts; the kind of person who can most recently be found wining and dining with the President of the United States and his coterie of MAGA sycophants. Wales, on the other hand, keeps things relatively low-key. Even as the site he founded, Wikipedia, turns 25 years old this month, he seems more interested in fixing his home Wi-Fi than joining the tech elite's performative power games. He has also spent the past few months promoting a new book, The Seven Rules of Trust, that uses Wikipedia's overarching strategy and unlikely rise to articulate Wales' playbook for fixing much of what's broken in today's deeply polarized and antagonistic society. On this week's episode of The Big Interview, Wales and I discussed what it means to build something used by billions of people that's not optimized for growth at all costs. During our discussion he reflected on Wikipedia's messy, human origins, the ways it's been targeted by governments from Russia to Saudi Arabia, and the challenges of holding the line on neutrality in an online ecosystem hostile to the notion that facts even exist. We also talked about what threatens Wikipedia now, from AI to conspiracy-pilled billionaires, and why he'll never edit an entry about Donald Trump. This interview has been edited for length and clarity. KATIE DRUMMOND: Jimmy Wales, welcome to The Big Interview. Thank you so much for being here. JIMMY WALES: Thanks for having me on. We always start these conversations with a few quick questions, like a little warmup for your brain. What's an internet rabbit hole you've fallen into most recently? People are working on extensions to deal with every kind of thing in the world, and it's amazing. What's a subject you never argue about online anymore? What do you trust more: Wikipedia or ChatGPT? What's your favorite website or app that is not Wikipedia? I really do like parts of Reddit. I lurk and read in the personal finance subreddit. I find that I spend a lot more time lurking in the Reddit app on my phone, because I would rather read thoughtful conversations than scroll on X. But all those things are in London. So if I lived in Silicon Valley, I would only have tech friends because that's who lives there. Whereas in London, it's much more comprehensive. They're kind of the same thing, but I think building trust is harder. Particularly these days, when we are in an era with such low trust that you have to get past people's skepticism. I have to admit, and this is very embarrassing for me because I'm Canadian, and I historically have liked to keep a low profile. So Jimmy, I have to ask you, what does that mean? What does it mean at this moment to have a Wikipedia page? Mainly it just means there's enough third-party, neutral, high-quality sources to be able to put together a biography and collate the information about you and so forth. But in general, you know, it is not that we have recognized you, it's that the world has somehow. One of the issues I'm now dealing with—I mean, this is a minor issue—is, sure, there are third-party sources, but those include small podcasts that I've talked to that a friend hosts where I share information about my spouse and my child, and then you realize, “Oh, that's now on Wikipedia.” I'd love it if it wasn't, but it's a lesson that you learn as you go. Have you experienced that with your own existence on Wikipedia? I think for a lot of people when you enter the public eye in some capacity, often you don't really realize it because you're on a small podcast with your friend and you're saying, “Oh yeah, whatever. No one's gonna listen to this.” But eventually they might. So yeah, I've had some funny experiences. One funny experience was when somebody vandalized my Wikipedia page to say that “in his spare time he enjoys playing chess with his friends,” which I love the sound of … Unfortunately, what happened is during a small gap of time before somebody took it out, because they didn't have a source, some let's just say … slightly lazy journalist actually took it from Wikipedia and put it in a biography magazine piece about me. So then there was a source, and I'm like, “Well, hold on a second.” So I contacted the journalist, and I'm like, “Where did you get that?” And they're like, “Well, it was in Wikipedia.” I'm like, “Well, it didn't have a source.” It is not there now, I don't think. I wonder why; maybe they read in a magazine that I'm a big chess fanatic. You've been doing a lot of press lately around the book. I was reading a bunch of interviews, and obviously Wikipedia is in the headlines often; I read in one of these interviews that you describe yourself as shy. I'm curious about how you handle being the face of Wikipedia, the author of a book, someone out there doing press. Actually, for the book, probably where this came up is that Dan Gardner, who I worked with on the book, it's thanks to him that we interviewed people, because he's like, “Yeah, we need to interview people.” Like, we interviewed Thomas Friedman and I'm like, “Well, I've known Tom for years and consider him a friend. I'm a little shy to ask him for an interview,” which sounds like nonsense. I really like working, I don't like working in a big office with lots of people and lots of meetings. I just wanna get on my computer in a quiet room like this one. Let me back up a little bit. You left your PhD program to get into tech. Was there a specific moment you remember in your academic time where you decided to leave? I was a research assistant for a top professor, and then he went off for the summer while I was trying to do my summer paper, which was a big deal. He wasn't there to help me. At the end of that process I was like, “I just really don't like this.” A friend who worked in Chicago said, “Well, why are you doing it if you don't like it?” And I said, “Well, I always wanted to work in the markets, but getting an undergraduate degree from the University of Alabama in finance, I couldn't get any good job.” And they said, “Well, come and talk to my boss,” because she worked in the financial markets. I did, and they offered me a job. It was an online encyclopedia with a much more traditional model: articles written by experts, a formal peer-review process. Can you talk a little bit about that origin story? I was describing Nupedia and how it worked, and there was a seven-stage review process to get anything published, and they said, “So that was the Seven Rules of Mistrust.” I said, “Actually, yes.” A whole system where you had to fax in your CV to prove you were qualified. In other words, there was that challenge of like, “Oh, I don't trust you.” You have to prove every step of the way. As opposed to the open, sort of friendly, “Hey, I'm doing this project, come and help me. Let's start writing together.” There was a lot to learn, and a lot didn't work because it was no fun whatsoever writing for Nupedia because it was very intimidating. There's a really fraught, and very moving, story in the book about your daughter, where you realize that digital publishing and the internet and access to information wasn't working as quickly as it needed to, which then led to this notion of Wikipedia, if I'm remembering correctly. I had been working on Nupedia for nearly two years, and it wasn't going well. We had, you know, a couple dozen articles done. In fact, one of the first articles that we published we had to retract, because it was found to contain plagiarism. As soon as we put it out, within three days somebody goes, “Actually, this huge passage is copied from somewhere else.” I was like, “Oh dear. Then, when Kira was born, she was very ill and in the hospital, and she had meconium aspiration syndrome and I met the doctor who was explaining it to me. Obviously, I rushed to the internet to try and research what this was, and it was pretty bad. You know, I could find a few blog posts and some academic literature, which was way above my head. I said, “Wow, we really do need an encyclopedia.” We've gotta do this.” I sort of ripped up the whole thing. I installed the Wiki software, which was an open source, very simple wiki package at the time. You could create an account, but your account wouldn't have a password, so anybody could log in as anybody. It wasn't even a finished product really, but I just thought we have to get started, because this is really important and life's too short. You know, when I think about Wikipedia and the Wikimedia Foundation more broadly, “scrappy” is a word that comes to mind. But today, the foundation has over 600 full-time employees across 54 countries. Your annual operating budget for 2025, 2026—when I say “your,” I'm referring to the foundation—$207.5 million in revenue against a similar number in expenses. I'm curious about the experience of actually scaling Wikipedia. There's a few things to say about that. So first, some 2 billion devices see Wikipedia every month. The other thing is there's two tracks that you can think about almost as being somewhat independent. That is the growth of the community and the website, which was very successful and did very well over a long period of time. People wanted us to do things, but I'm like, well, there's just me here and two other people. I'm not talking about anything exotic or unique to Wikipedia. You make a few bad hires early on and you're like, “Oh, that didn't work out.” It took some time to think about professionalizing the foundation, while at the same time keeping that real community spirit. A lot of the foundation staff are Wikipedians first. They come to things through being well-known on Wikipedia. Other people come in from various expertise, so that's always a challenge. But there are companies that go from zero to 12,000 in just a few years, and it's like, wow. That's gotta be a complete nightmare. One of the unique challenges to Wikipedia in a global context is that if you're operating in 54 countries, you're going to be running into regimes around the world, authoritarian governments, who want to control the flow of information that you're providing in ways that Americans are not accustomed to. It is something that we've dealt with for years, and I assume we always will, but it does seem to be a little more difficult more recently, as we see increased legal problems in different places that are clearly politically motivated. We're under threat of block in various places. We are completely blocked in China at the moment. Then obviously we have volunteers who are sometimes at risk. We had a volunteer in Belarus who was arrested. We've had a couple of volunteers in Saudi Arabia arrested. These things are heavy, and we really take them seriously. It is true, whatever's going on politically in the US, which is a lot, we still have the First Amendment. Ironically, at the same time they're canceling the visas of people who they politically don't agree with, they're also sort of campaigning that Europe shouldn't censor people's speech. So they're a mixed bag on that front. The idea for most Americans that somebody is engaging in a dangerous and heroic act, just to write down that Russia invaded Ukraine, is really something to get your head around. I think it bears repeating that Wikipedia editors are volunteers. They are doing this in their spare time. What are the limits of your ability to protect them in these contexts? We've had cases in the past where it's really hard. Sometimes volunteers would come to me and say, “Oh, why aren't you writing a statement to The New York Times?” I was like, “Well, actually, behind the scenes, one of the problems is that sometimes the families prefer to just lay low.” If you make a big deal out of something the person who might have been in prison for six months could have it arbitrarily extended to 10 years—or be executed. What we try to do is work with the right professionals, like human rights lawyers. But the truth is, in many cases, it's a genuine risk that people are facing. One of the important things that people can and should do is be very, very careful about their privacy online. Use a VPN, really take the technical measures to protect yourself, because that's probably as effective, probably more effective, than what anybody can do for you after the fact, depending on where you are and what's going on, obviously. Does that advice feel more salient in the United States than it did a couple years ago? But in one interview I started talking about something and the interviewer said, “Jimmy, you're not cheering me up. You're supposed to be the optimist here.” I was like, “I don't think Trump is gonna have a coup, or engage in a coup. But you might as well use that VPN. That's one example in the technology world where—I mean, I live in the UK and VPNs are under political attack. Literally everybody should be using a VPN. I want to ask you a little bit about bias. It's very much a core principle that informs how Wikipedia is run, how it's edited, what ends up sticking to the page and what doesn't. Whether it's Wikipedia or a newspaper, there is a really important element of trust to non-biased publication of information, right? There has been, and continues to be, especially in the media, a very heated debate about the idea of objectivity, about the idea of unbiased publishing and whether that's even possible. It sounds like you certainly think that it is, and I'm curious for you to share how that actually works for someone to sit down and edit a Wikipedia entry. Maybe they themselves have very strong feelings about Ukraine, about abortion, about whatever the subject is, but how does setting that aside actually work in practice? If somebody says to me, “Oh, I don't think neutrality is even possible. I don't think there's such a thing.” I say, “But wait, do you think there's such a thing as bias? If you really feel super emotional about a particular topic, it's going be hard to be a good Wikipedia writer on that topic. I wouldn't edit anything about Donald Trump, because the man makes me insane. But at the same time, I do think it's possible. They're going to work together writing about abortion. They both understand Wikipedians can't take a side, can't say abortion is a sin. One is: How emotional you are about it. That can be a problem if you can't calm yourself down enough. But in other cases, how ideological you are might not inhibit you from writing in a fair way. I mean, that Catholic priest and that Planned Parenthood activist may never agree, and they're not about to budge in their views, but they both probably believe if you understood the issue as well as they do, you would probably agree with them. You see this a lot in social media debates. I think they're just not that confident in their ideas, because if they were that confident—like if somebody said to you two plus two equals five, you wouldn't fly off the handle and go insane at them because you would be like, “Oh, that's a very odd thing to say. And certainly if that were true, planes would crash. That's a completely bonkers thing to say.” But we don't get mad about it. Again, I don't think many people are very angry at people who think the Earth is flat. You just think, “Can you tell me more, because I'm really interested in where in the world did you come up with such an idea?” Whereas if it's something that we're less confident about, that's where we can get emotional and so forth. One of the criticisms leveled at Wikipedia has been that Wikipedia relies a lot on the press. If an individual thinks well, all of those sources are leftist, they are biased, therefore Wikipedia has an inherent problem with bias because of what it relies on. How do you see the media at this moment? What would you like to see, if anything, change about how journalists, including WIRED, approach politics, approach abortion, approach all of these subjects that are so fraught and that are really tied up in this conversation about trust that is so fundamental? There's so many elements to pick apart. We could probably have a four-hour conversation about just this, but the accusation that the media is vaguely liberal is probably not entirely untrue. Hopefully, in my view, at really good publications, which is a lot of publications, the journalists may vote a certain way, but they take the time to be good Wikipedians themselves and to report in a neutral way. You always have to be careful about that. There was some claim that now Wikipedia has banned all but left-leaning sources. I'm like, “Well, yeah, if you think The Wall Street Journal is a communist rag, then maybe that's true.” But that actually isn't true. I think there has clearly been a rise in low-quality, populist right-wing media, and we could probably name a few of the famous names. What I would say is, if there are any right-leaning billionaires who are worried about the media, why don't you fund some really good, quality conservative journalism? If you think The New York Times is sort of vaguely liberal, have another newspaper that's sort of vaguely conservative. What are our processes that aren't working well?” Or, you know, maybe sometimes we look into them and we're like, actually everything's fine. I'm sorry, we're not going to treat the comments of a social media influencer on vaccines the same as the New England Journal of Medicine. At the same time, let's be careful that we aren't jumping on a bandwagon and in a moment of emotion. Let's try to be the place where we say “here's the parameters of the debate” rather than trying to take sides in that debate. One of the challenges I have come up against in my job around neutrality and bias—actually, it came to mind because I was reading an interview you did with The New York Times–where you said, “In various places around the world, not speaking just of the US, facts are threatening. And if your policies are at odds with the facts, then you may find it very uncomfortable for people to simply explain the facts.” A lot of people in the United States reject factual information. They might level a term like bias when in fact they're talking about something that is just true. I've resigned myself, in some cases, to just accepting that those people live in a different version of reality than I do, and it is a reality informed by certain fictions. How do you think about that, and how do you think about reaching those people? So much of your book is about trying to reach more people with this shared framework of trust. That is definitely hard, but I actually think there are fewer people who are really hardcore about that. I think most people on all political spectrums will say, “I want to be told the truth. I want to make decisions based on facts. I don't wanna live in a delusion.” They may be rejecting elements of the media or this, that, and the other, sometimes for bad reasons, sometimes for good reasons. An old friend from a million years ago said something on Facebook and I was like, Hmm, hold on. But usually, the response might've been, “Well you do know, of course, The New York Times is left-leaning and they may have left some things out that might be uncomfortable for their position, you should read this.” That is a conversation; that's something we can work on. But just rejecting it wholesale, as if the mainstream media is propaganda? I think Elon Musk has talked about how Wikipedia echoes mainstream propaganda. It's a really odd thing to say. For the media, I think it's really important to be aware that those people exist, and a few of them are just not reachable. I assume that's been true throughout all human history. But there are a lot of people who are like, “No, actually, I'm not listening because you've made these mistakes in the past and didn't correct it.” I live in the UK, and one of the big political things that we've been through was Brexit. The people who were for Brexit were not persuaded by that, because calling people racist doesn't usually change their minds about anything. They were like, “Hold on, you're not listening.” In fact, people probably weren't listening, and that led to a protest vote that actually had dire consequences. I was gonna say, yeah, look at you now. At the same time, these things happen. One of the symptoms of all this, that I have been pointing out recently, is the US just went through the longest-in-history government shutdown, because Congress is so filled with partisan hacks. They are unable to trust each other enough to find some bipartisan solutions and some compromise. You mentioned Elon Musk a few minutes ago, and this is someone who has built an alternate-reality version of Wikipedia. Do you know why he is so opposed to Wikipedia's continued existence? Wikipedia feels threatened in so many different ways, right? There's the Elon Musk of it all. How precarious do things feel for Wikipedia right now, and what feels like the most pressing threat to you? Are you waking up at 3 in the morning because of Wikipedia? Are you waking up at 3 in the morning because of AI? What are you thinking about Wikipedia's place in the world right now? It doesn't even come across my mind most days. One of the things I say in the book but I've also just said to Elon, is that putting out this idea that Wikipedia has been taken over by far-left activists has two harmful consequences. One, you're telling those kind and thoughtful conservatives who might help with whatever bias you see that Wikipedia is going to be a hostile place for them. And you're telling those crazy woke activists that Wikipedia is their new home. Then we have to deal with them. So for me, the thing I worry about is community health. Are we treating people well when they come in? In this era of hostility and these culture wars, are we at risk of letting those cultural wars infect our culture of neutrality and being thoughtful and being kind. The other thing that's funny about Wikipedia is we are a charity, we're a nonprofit organization. As such, when we think about the competitive landscape in a kind of dotcom way, “Ooh, what if Grok takes 30 percent of your traffic? What's that gonna do?” I'm sure it'll hurt the foundation's revenue if we have less traffic. But is it gonna change anything about the community? So we'll carry on, even if at a different scale. But also I think as long as we keep our spirit, I think people will always really want that sort of information that's not just written by human, but that's also been chewed on and thought about and debated and worked out by humans to say, “How can we make it as good as we possibly can?” What would you tell someone listening who loves Wikipedia, who uses it all the time, besides donating, what can they do to make sure that Wikipedia is around and thriving in 10 years? I always encourage people, if you like the spirit of Wikipedia just make a little edit now and then. Does anyone ever tell you that the idea of becoming an editor on Wikipedia is intimidating? That it feels a bit too powerful? Just pick something obscure and make a little edit. If you've read a book and you're like, oh, actually there's not that much about this book, let me add a little something. I remember writing high school papers and being told, “You cannot use Wikipedia as a source.” But just looking at everything coming out of Venezuela the last few days and seeing how much of it was AI-generated, Wikipedia now feels much more rigorous and accurate than so much of what we're seeing online. Where do you see our digital experience going from here, and what do you see as Wikipedia's place in that? I think we all have to worry about it, not just in the context of Wikipedia, but more broadly in the decline of local newspapers. Because I like to be optimistic, I was thinking the other day, oh, are there ways that AI could somehow support local journalism? That's a really fraught topic because the easiest way is the lazy way, which is just turning AI loose, starting to write a bunch of nonsense, and it's just gonna be slop. They asked whether you were “the last decent tech baron,” which is a very flattering way to be described. When I think about a tech baron, I think about Mark Zuckerberg feeding his cows macadamia nuts on his compound in Hawaii. I don't know that I would necessarily think of you as a baron, but I'm curious about how you think about yourself relative to the Mark Zuckerbergs and Jeff Bezoses and Sam Altmans of the world. But the way I live my life is quite different. At the same time—and this is where I can be a little bit grandiose—in 500 years, people are gonna look back at Wikipedia and go, you know what? That was one of the good things. That was a really difficult time in human history, and there was a lot of crap, but a bunch of really nice people got together to gather knowledge and share it with the world. I'm really happy and proud to be a part of that. What advice would you give the tech barons about how they can participate in rebuilding trust? So many people don't trust what those barons have built. They don't look at Facebook and think that is a place to find trustworthy information, a place to have trusted conversations. They don't look to the AI industry and feel a sense of trust in those leaders. I think a big part of it is the algorithms that are optimized for short-term engagement. They have such clear negatives associated with them that you should really rethink that as a business strategy. I've said this to people at Facebook. If people become convinced that you are part of the destruction of civilization, that's not good for business in the long run. Optimizing for next quarter, sure, you could probably goose your revenues a bit, but it isn't really the right answer. There's more important things that you could do. As a serial optimist, do you think we are too far gone? My best evidence for that is to step back and not think about social media for a minute, and just think about people in your personal life and the way we treat each other. When you step into an elevator you don't immediately think, “OK, which one of these people's gonna try to slit my throat?” The elevator gets stuck halfway between floors. They're people in an elevator just like you. We see that and we trust people all the time in all kinds of ways. I think we can just say politics has gotten super broken. So, what piece of technology would you love to control? What would you love to alt, so alter or change, and what would you love to delete? I would love to be able to control my home Wi-Fi network, which has been driving me crazy all morning. Broadly I would say the algorithms of social media. I've said for a long time, if I went on Facebook one morning and they gave me an option to say, “We'd like to show you things that we think you're gonna be less interested in but that we think are high quality. You may not agree with them, but we think they're high-quality because we've got these signals. Do you wanna see more like that?” I would say yes. But what would it do to their ad business? Well, maybe in the long run I would revert a long trend, which is that I hardly ever go on Facebook anymore, you know? I deleted my account maybe two years ago. I can't even think about it. I cannot disagree with you there. I think a lot of people, a lot of journalists, feel like they need it professionally. Because you can post stuff there and people will see it. Actually, one of my good friends quit, and I was like, “Well, OK, there's one more reason to just not bother.” One more reason to give it up. I think they should just delete it and start over. Maybe start over with a different owner. That would be a good start, yeah. You can always listen to this week's podcast through the audio player on this page, but if you want to subscribe for free to get every episode, here's how: If you're on an iPhone or iPad, open the app called Podcasts, or just tap this link. You can also download an app like Overcast or Pocket Casts and search for “uncanny valley.” We're on Spotify too. Big Story: Understanding Trump's retro coup in Venezuela Billion-dollar data centers are taking over the world Livestream AMA: Welcome to the Chinese century WIRED may earn a portion of sales from products that are purchased through our site as part of our Affiliate Partnerships with retailers. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of Condé Nast.

Shortage drives fraudsters to target legacy GPUs When you purchase through links on our site, we may earn an affiliate commission. The ongoing DRAM shortage has driven the best graphics card prices to new heights for gamers and also created a perfect storm for fraud. In a revealing YouTube video, Brother Zhang, a renowned graphics card technician in China, exposes how scammers are using the old switcheroo tactic on Nvidia's last-generation GeForce RTX 40-series (codename Ada Lovelace) graphics cards to scam unsuspecting buyers in the second-hand market. In one case, the graphics card in question was the Gigabyte GeForce RTX 4080 16GB Aero OC, a custom variant of the GeForce RTX 4080 that retailed for around $1,199 at launch. The seller claimed the graphics card was not functional and wanted to offload it for just $143.50 on Xianyu, a popular second-hand marketplace in China. The suspiciously low price, approximately 12% of the graphics card's original retail value, would raise a red flag among potential buyers. It would be an opportunity of a lifetime for anyone with repair skills or access to a skilled technician like Brother Zhang. Nonetheless, the graphics card is still salvageable because you can transplant the silicon and memory chips to a new PCB, provided they are legitimate and functional. The bad news just kept pouring in as Brother Zhang continued his inspection. It doesn't surprise us, as Chinese factories were repurposing these mobile Ampere dies into desktop graphics cards a couple of years ago. The sham didn't stop there, though. The GDDR6X memory chips were also highly suspicious. Brother Zhang suspected they were either fake, defective, or salvaged chips from donor graphics cards. Still, realistically, it's only a matter of time before it has to resort to a more significant price increase. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Graphics card scams have always been around, and they flourish during hard times, such as the COVID era or the cryptocurrency mining boom. If you're considering purchasing a second-hand graphics card from Facebook Marketplace or eBay to save money during this shortage, do your due diligence and proceed with extreme caution. Even supposedly “new” graphics cards from questionable sellers have turned out to be elaborate swindles nowadays. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds. Zhiye Liu is a news editor, memory reviewer, and SSD tester at Tom's Hardware. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher. © Future US, Inc. Full 7th Floor, 130 West 42nd Street, New York,

When you purchase through links on our site, we may earn an affiliate commission. Elon Musk has confirmed on X that Tesla has restarted work on the Dojo3 supercomputer following the new success of its AI5 chip design. Musk also added that he is hiring more people to help build the chips that will inevitably be used in Tesla's next-gen supercomputer. This news follows Tesla's decision that it was cancelling Dojo's wafer-level processor initiative in late 2025. Dojo 3 has gone through several iterations since Elon Musk first chimed in on the project, but according to Musk's latest thoughts on it, Dojo 3 will be the first Tesla-built supercomputer to take advantage of purely in-house hardware only. Now that the AI5 chip design is in good shape, Tesla will restart work on Dojo3. AI5 chip will be roughly comparable to a Nvidia Hopper Chip in a single SoC, and Blackwell class when you run two of them together.Those AI chips from Nvidia run ~$25K-$50K each. Work on Dojo 3 coincides directly with Musk's new nine-month release cycle, where Tesla will start producing new chips every nine months, starting with its AI6 chip. AI7, we believe, will likely be an iterative upgrade to AI6; building a brand new architecture every 9 months would be extremely difficult, if not impossible. Dojo 1 was supposed to be one of the most powerful supercomputers when it was built, but competition from Nvidia prevented that from happening, among other problems. If Tesla can deliver competitive performance with Nvidia GPUs consistently, Dojo 3 has the potential to be Tesla's first truly successful supercomputer. Elon also hinted that Dojo 3 will be used for "space-based AI compute". AI4 by itself will achieve self-driving safety levels very far above human. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Aaron Klotz is a contributing writer for Tom's Hardware, covering news related to computer hardware such as CPUs, and graphics cards. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher.

Human-bear encounters are up 163% since 2021, with more than 12 human deaths recorded in 2025. When you purchase through links on our site, we may earn an affiliate commission. A Japanese city is rolling out an anti-bear drone system in response to “record levels of bear casualties and sightings.” Terra Drone has announced (machine translation) that its bear-repelling drones will be deployed around Ishinomaki City in Miyagi Prefecture to address concerns regarding a “bear infestation.” A number of factors are thought to have resulted in the current situation. Meanwhile, city developers increasingly move nearer to bear habitats, closing the circle on their free roaming areas. It is also claimed that the police and similar agencies face constraints on lethal training to address wild animals. The resulting drone system can “combine speed, safety, and immediate effectiveness.” In action, the drones are operated from up to a kilometer (~1,100 yards) away, ensuring operator safety. A drone can deliver its capsaicin spray on target at distances between 5 and 10 meters (up to ~11 yards) from the bear. This has a temporary, non-lethal impact on the bear(s), and makes sure they can't sense humans in their vicinity. Ishinomaki City is far from the only area in Japan affected by bears. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Last year, we reported on Japanese tech giant NTT's design for a laser drone system designed to protect chickens. It is good to hear and see advanced drone systems being used outside the theater of war. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds. Tom's Hardware is part of Future US Inc, an international media group and leading digital publisher.

Three little girls were born with modifications to their genomes that were intended to protect them against HIV. The changes he'd made to their DNA were permanent and heritable, meaning they could be passed down to future generations. A Chinese court sent him to prison for three years, and the Chinese government banned genome editing for reproductive purposes. Since his release in 2022, He says, he's worked on a gene therapy for boys with Duchenne muscular dystrophy. He has yet to publish or share any results publicly, but he claims that a pharmaceutical company has taken on his Duchenne research and that funders are eager to help him continue his work. And He, who has set up an independent lab in south Beijing, recently started talking again about human embryo editing—this time to prevent Alzheimer's. With germline editing prohibited in nearly every country including the United States, his path forward is unclear. Through it all, He has documented his life on social media. He has posted about his failed romance with self-styled “biotech Barbie” Cathy Tie, a Canadian former Thiel fellow and cofounder of a human embryo editing startup. A condition of this interview was that WIRED refer to He as a “pioneer of gene editing,” but he has more colorfully referred to himself on X as “Chinese Darwin,” “Oppenheimer in China,” and “China's Frankenstein.” Here are 23 ways China is rewiring the future. He often posts photos of himself in a crisp lab coat, posing alone near scientific equipment. One glaringly empty lab shot comes with the text “I did not violate ethics, I overturned it.” More recently he dropped the austere look and posted an image of himself seated on a giant throne with prehistoric animals at his feet, a rainbow beaming down on his crown, and a double helix adorning his purple robe. WIRED spoke with He about designer babies—the ones already born and the ones he hopes to eventually produce. This interview has been edited for length and clarity. Of course not, but they made history. I'm lucky that Lulu, Nana, and the third girl were healthy; they're normal. We have observed them for seven, eight years now. We should give a trial to maybe 300 now. Their family is very happy with it. Have their parents told them that they were gene-edited? What is your new lab focusing on? People with this mutation are free of Alzheimer's and even live longer. Are you currently working with human embryos? For now, we're not getting into the embryo yet. We're still on human cell line experiments. And you've been doing this for the past couple of months? Because it is going to change the world. Donations and some people who provided early seed money. Maybe every week, an investor comes to Beijing, wants to have dinner with me and also wants to invest in me, or maybe they propose some business ideas. How much has your lab raised so far? You mentioned on X that a patient had donated some money. Oh, yes, the patient is an entrepreneur of a big company. Is it fair to say that your funders are wealthy people who want their own babies to have protective traits? How many people are working in your lab on Alzheimer's? It's just me plus two, three people. Are you allowed to carry out a clinical trial in China given your criminal history? Germline gene editing is currently banned in China. I wanted to start a lab in Austin, Texas, to continue the embryo editing research, but I don't have a passport. But for the human trial, it has to be somewhere else. South Africa made it legal last year. [Editor's note: In August 2025, South Africa removed language from its national ethics guidelines that some had interpreted as potentially allowing heritable human genome editing.] I already talked to people from South Africa. Are you allowed to leave the country? You have condemned research on IQ enhancements, but some people might consider editing embryos to prevent Alzheimer's a form of enhancement. Where do you draw the line between disease prevention and enhancement? I think that is a Nazi eugenic experiment. The scientists working on this should be arrested if they want to enhance human IQ for the billionaires. With Alzheimer's or HIV, it's just for preventing human disease and never to enhance the IQ of a person. How many years away do you think you are from trying to create babies that have been edited for Alzheimer's protection? And then we wait for regulatory approval in South Africa. There are already animals being created from highly edited embryos. Like the dire wolves that were revealed in 2025 and the pigs that are being used for human transplants. Do you think it might be possible to one day do multiple genetic edits to a human embryo? That's exactly what we are developing in our lab. If we could have multiple edits that could prevent cancer, Alzheimer's, cardiovascular disease, HIV, and others, in a panel of 10 to 20 genes, that will prevent all the major diseases that are happening today. The children born will be much healthier and maybe even live longer than us. I could see this research being controversial to some people. How do you plan to avoid run-ins with Chinese authorities? Just be open, transparent, do the right thing. Unfortunately, it's not open for foreigners anymore due to national security reasons. I'm not going to go against the law. You gained instant fame with the birth of those babies and then spent time in prison, and now here you are continuing your work. What have you learned in this time? Make sure that it improves their health or happiness. That's the number one requirement for any medical research. You don't worry about ethics, national geopolitics, or anything. Alternatively, you can submit a letter to the editor at mail@wired.com. Here are 23 ways China is rewiring the future Why your next robot coworker will probably be from China Trump declared a moon race with China. China's Gen Z women are driving an AI boyfriend boom You've never heard of China's greatest sci-fi novel In your inbox: The biggest tech news coming out of China WIRED may earn a portion of sales from products that are purchased through our site as part of our Affiliate Partnerships with retailers. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of Condé Nast.

When DeepSeek burst onto the global stage in January 2025, it seemed to appear out of nowhere. The country's top internet regulator, the Cyberspace Administration of China (CAC), requires that any company launching an AI tool with “public opinion properties or social mobilization capabilities" first file it in a public database: the algorithm registry. In a submission, developers must show how their products avoid 31 categories of risk, from age and gender discrimination to psychological harm to “violating core socialist values.” Only then is a tool publicly listed in the algorithm registry. While the European Union is pursuing a single, comprehensive AI Act, notes Matt Sheehan, a research scholar at the Carnegie Endowment for International Peace, China's approach to regulation is more ad hoc, targeting specific algorithms and building up iterative standards. (The US has no comparable registration system or centralized regulatory agency.) *Data current as of April 2025, includes both “generative AI” and “deep synthesis” algorithms One assists ob-gyns in a Shanghai maternity ward; another helps manage state power grids. Kendra Schaefer and her colleagues at Trivium China, a Beijing-based policy consultancy, have been compiling the CAC's updates into a comprehensive database, enriched with their own research. Each city has its strengths: Beijing's elite universities, national labs, and political strength give it an edge in large-scale innovation; Shenzhen (in Guangdong) is home to a dense hardware supply chain and vast pool of engineering talent; Shanghai, close to multinationals, excels at commercialization; and Hangzhou (in Zhejiang) is fueled by Alibaba's ecommerce empire. Chongqing is positioning itself as an AI manufacturing and logistics node; and heavy state investment has helped Hefei, in Anhui Province, become known as “China's speech valley” for its cluster of speech-recognition firms, including iFlyTek. Filings also originate in less obvious regions like Guizhou, China's “Big Data Valley,” where massive data centers power Huawei's Pangu model, and Inner Mongolia, where state enterprises are integrating AI into mining and agriculture. In the Trivium dataset, state-linked listings—from state-owned enterprises to government-backed research institutes—make up 22 percent of filings. Many state-linked firms partner with Big Tech to build their AI: PetroChina, for example, teamed up with Huawei and iFlyTek to create oil and gas applications; State Grid used DeepSeek to build a model optimizing power grids. Ikea, for example, has a smart shopper algorithm that generates product recommendations. Yum China, the parent company that operates Kentucky Fried Chicken in China, listed a model that generates menus and promotional material. Unlike in the US, where OpenAI, Anthropic, and Google DeepMind dominate the market, China's competition to build foundational AI remains diverse and contested. China's six “AI tigers”—Moonshot, Minimax, Zhipu, Baichuan, 0.1AI, and Stepfun—are all backed by Alibaba or Tencent. ByteDance's Doubao surpassed DeepSeek as China's most popular chatbot, but its spot at the top is not assured. While the giants duke it out for chatbot supremacy, startups are hard at work in every sector imaginable. Founder Derek Li says his 12-year-old company is leaps beyond the ed-tech competition. They “put wheels on a horse,” he says, bolting AI onto their existing stale software. Squirrel claims to diagnose knowledge gaps, measure progress, and adjust lessons in real time. When China banned for-profit tutoring in 2021, the company's revenues collapsed overnight. Squirrel's network includes more than 3,000 centers across China, serving 1.2 million students. Li, who withdrew his sons from a private school in Shanghai so that they could be home-schooled on Squirrel's platform, says that “in the future, teachers won't teach knowledge.” Instead, he says, “they'll become data analysts, understanding learning reports and students' ability, and psychologists, understanding emotions and shaping their personalities.” AI Kanshe (translated as “AI Sees Tongue”) is a traditional Chinese medicine startup that analyzes health through images of the tongue, palms, and face. A longtime student of tongue and hand diagnosis, Li wanted to combine the diagnostic methods of traditional Chinese medicine with modern machine vision. Its model is trained on more than 100,000 annotated images of tongues, hands, and faces. Founded in 2024 by Wu Song, a former Wall Street quant trader, Zhongtan Puhui Cloud Technology develops AI-driven tools for carbon accounting. The green transition, Wu says, still relies on cumbersome human labor that could be automated. Based in Shenzhen, the company now produces robots for factories, schools, health care facilities, and homes. The company also develops robots for elder care and companionship, as well as consumer devices like smart vacuums and lawn mowers. The company currently sells to more than 40 countries, with roughly a third of its revenue coming from overseas markets. The company was founded by former Tencent vice president Yao Xing. After leading Tencent to develop a Go-playing AI, Yao concluded that AGI progresses fastest when its goals are clear and measurable—as in the game of Go—but that most real-world challenges are open-ended. So in 2020, he left Tencent to found XVerse, aiming to build complex, simulated worlds where AI could learn, act, and evolve. In 2024, the company launched its flagship product, Saylo, which turns chat conversations into immersive AI-driven story videos. Users create personalized virtual characters and explore branching plots in real time. The company also built its own large-language model, XVerse MoE, trained on diverse data sources including news, code, literature, and academic papers. The game evolves dynamically based on players' decisions, creating endless possible plots and endings. BubblePal, a golf-ball-sized orb, transforms children's toys into chatty companions. A companion app lets parents set learning goals, such as encouraging more English conversation, and track weekly “growth reports” that analyze a child's conversations, emotions, and interests. Since its launch last summer, more than 300,000 units have been sold. The toy is made by Haivivi, a Chinese company founded in 2021 by Li Yong, a former Alibaba executive who helped develop the company's smart speaker. Realizing that many of the smart speaker's users were children, Li decided to make children's toys, and in 2023 pivoted to AI companions. Haivivi aims to add more popular characters such as Lucky Cat and develop product lines for young adults, like AI-powered tarot cards. In recent years, the term chuhai—“to go overseas”—has become a buzzword as Chinese firms look abroad to escape slowing consumption and weak capital markets back home. A top earner for the Hangzhou company Glority, for example, is a plant-identification app called PictureThis, which is popular outside China. To avoid geopolitical trip wires, some Chinese tech companies have been distancing themselves from their roots by hiring foreign staff, setting up headquarters in Singapore or California, or leaving China altogether. Shortly after the US established export controls on Nvidia chips, the company relocated to Singapore, laid off most of its Chinese staff, and scrubbed all of its content from Chinese social media. Alternatively, you can submit a letter to the editor at mail@wired.com. Why your next robot coworker will probably be from China China's Gen Z women are driving an AI boyfriend boom In your inbox: The biggest tech news coming out of China WIRED may earn a portion of sales from products that are purchased through our site as part of our Affiliate Partnerships with retailers. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of Condé Nast.

There's a particular kind of sci-fi nerd who equates fusion tech with utopia. If we could only harness the engine of the stars, it would uncork near limitless energy and neatly sweep away a whole mess of humanity's problems. The Chinese solar supply chain can now pump out 1 terawatt of panels every year. In China itself, vast energy megabases combining solar and wind stretch for miles in the country's western deserts and Tibetan highlands, each producing the power of multiple nuclear plants and connecting to population centers in the country's east via ultrahigh-voltage power lines. Huge factories, urban apartment buildings, and humble village homes are plastered with panels. In Europe, Chinese-made photovoltaic panels are so cheap that they cost less than fencing materials. By now, major headlines have begun to catch on to the reality that China's renewable energy revolution is one of the biggest stories in the world, while Donald Trump's anti-renewable vision of American energy dominance is a backward sideshow by comparison. But chroniclers of this green tech revolution almost always understate its chaos. At this point, it is far less a tightly managed, top-down creation of state subsidies than a runaway train of competition. It is a panorama of coal communities decimated, price wars sweeping across one market after another, and electrical grids destabilizing as they become more central to the energy system. And absolutely no one—least of all some monolithic “China” at the control switch—knows how to deal with its repercussions. In the United States, 2024 was a record-breaking year for solar. Across the entire country, those 12 months saw some 50 gigawatts of new solar capacity added. Now consider some different numbers, for scale and contrast. The reason for this brain-warping mad dash of solar development? At the start of 2025—in an attempt to rein in the renewables sector—Beijing announced that it would discontinue a long-standing policy that had effectively propped up renewable energy prices, pegging them to that of the “baseline” coal power in each province. Any solar capacity that went in after May 2025, Beijing declared, would no longer get this deal. After May, sure enough, new solar deployments plummeted. The ensuing four months each added just 10 gigawatts of new solar on average, half of the prior year's pace—but that's still considerably faster than America at its peak. In China, one problem with all this burgeoning, majestic new solar is that it's completely overwhelming the national electrical grid, technically and economically. For electricity markets to work, grid managers must constantly balance supply and demand—but the former can't always be throttled back when it exceeds the latter. And some Chinese coal plants provide heat to communities through steam—so they need to run even if the electricity they generate is superfluous. One perverse result of all this energetic over-supply is that a lot of solar power simply gets wasted, or “curtailed,” to make way for dirty forms of energy that are harder to turn off. Another is that the inherently intermittent power of renewables simply makes it more challenging for managers to keep the grid stable. In August 2024, in China's far western region of Xinjiang—where the renewables build-out is at its most grandiose—poorly handled voltage fluctuations from solar and wind caused a regional blackout and even threatened the national electrical system, according to the South China Morning Post. As Econ 101 teaches, prices go down when supply rises faster than demand. But in most markets, there's an end point to this process: free. This, combined with the absolute imperative to keep the grid balanced, can create negative prices, which have become common in China's heavily populated Shandong Province. Decades ago, the metals giant Weiqiao Aluminum left the Shandong grid in favor of running its own captive coal fleet to power its smelters; this past year it plugged back into the grid to take advantage of cheaper rates coming from green tech. What's more, Chinese solar manufacturers—who, if you'll remember, might just be saving the world—are not even making money for their troubles. Oversupply of this product has caused prices and profits to collapse. But so far, this seems to be a long shot. The story is similar higher up the supply chain: Manufacturing capacity to produce solar ingots, wafers, and panels exceeds demand, sparking profit-killing price wars as firms try to keep their market share. Meanwhile, rapid technological innovation keeps forcing firms to invest in expansion with the latest and greatest designs, lest they get left behind by their competitors. When a completely new breed of solar panels yields a 10 percent difference in generation capacity, manufacturers who fail to quickly set up new production lines do so at their own peril. For the consumer, this added capacity means more energy with less real estate, perhaps a smaller parcel of land to purchase for your solar farm. Around 2022, a global spike in natural gas prices made the Pakistani electrical grid even more expensive and less reliable than usual. But instead of just suffering or firing up diesel generators, millions of Pakistanis installed solar panels to free themselves from the grid. The country imported so many Chinese solar panels that the grid as a whole began to fall into what is called a death spiral. Customers started opting out, leaving the grid to charge ever higher prices, which led even more people to flee, and so on. So Chinese firms are undermining the economics of major development projects funded by state-owned banks. If only there was more electrical storage—technology that holds solar power generated during the day and releases it in the evening—then a lot of the issues that torture China's renewables market would be resolved. Solar panels would become more valuable to the grid, their generation wouldn't have to be curtailed, and the producers would be able to sell more of their products at better prices. Of course, China has come to dominate this fast-growing sector as well: It is by far the world's largest battery maker. But China's electrical system hasn't figured out the rules and pricing to push battery capacity onto the grid fast enough to keep up. And besides, the vast majority of the batteries that China is churning out do not end up in grid storage. They're revolutionizing another increasingly Chinese-dominated green industry that is going fast, cheap, and out of control: automobiles. In 2018, the city of Shanghai lured Tesla in to build a gigafactory with what seemed like an especially attractive offer. For years, foreign automakers like Ford, GM, Volkswagen, and Toyota had dominated the Chinese car market—the world's largest—but were required to form joint ventures with Chinese firms for the privilege of doing business in the country. Shanghai told Tesla it could fully own its Chinese operations. Subsidies for land and cheap loans were proffered as well. As with solar, a wave of entrants battled for market share, causing profits to evaporate but giving consumers great choices at excellent prices. By 2024, nearly half of all cars sold in China had plugs. Internal combustion automakers in China are suffering or shuttering. And the reverberations are perhaps even stronger abroad. China shifted from an also-ran to the world's dominant auto exporter over the past five years, displacing Japan, South Korea, and Germany. Other countries have begun to panic that their own auto sectors could be eviscerated by the competition of cheaper, cleaner vehicles from China. Yet just like in the solar sector, where thin profits and high debts have turned some firms into sprinting zombies, the Chinese EV space is full of failed and failing enterprises. Its sales over the past few months have faltered badly compared with the previous year, and murmurs about its debt burden lead some to wonder if it too might collapse—even as people around the world are still dying to get their hands on these cars, and other automakers strain to compete. In sun-blessed Australia, where rooftop solar panels sit atop nearly a third of all households, the country's energy minister, Chris Bowen, proposed a “solar sharer program” to offer three hours of free electricity on sunny days. Solar and battery systems have allowed Hawaii to close its final coal power plant, and such systems are similarly helping other islands like Jamaica to reduce their need for imported fossil fuels. Donald Trump hates many people and things, but wind turbines and solar panels seem to hold a special place of contempt in his heart. His administration has attempted to cancel major offshore and onshore wind projects, along with plans for Esmerelda 7, a solar megabase slotted for the Nevada desert that would have been worthy of Western China. Trump and his energy secretary, Chris Wright, often speak of American energy dominance, but they are crippling American firms' ability to deploy and build the cheapest sources of electricity in the history of this planet, in favor of a combination of long-in-the-tooth arguments about fossil inevitability and long-shot bets on small modular nuclear reactors and, yes, fusion. Even among billionaires who don't share Trump's belief that climate change is a hoax, this latter affinity for far-out, breakthrough technologies has long been a hallmark of American climate investment and philanthropy. This attitude is epitomized by Bill Gates, who once dismissed existing green technologies like solar and wind power as “cute.” Instead, Gates has always preferred a lordly, capital-intensive variety of decarbonization, plowing dollars into sci-fi technologies that remain in a perpetual state of being just five years away—not the rapid, messy approach involving solar panels sprouting on every rooftop and recalibrating electricity pricing schemes. The green tech revolution—whose violence is principally financial, a withering assault on the value of fossil firms' assets—is not a dinner party. It could still be held back or slowed down. And it's a good thing, too, because there is another force powered by the sun's fusion that is also arriving at a force and scale that we are not prepared for: climate change. Through all the chaos, that system is getting a major upgrade. Let us know what you think about this article. Submit a letter to the editor at mail@wired.com. Why your next robot coworker will probably be from China China's Gen Z women are driving an AI boyfriend boom You've never heard of China's greatest sci-fi novel In your inbox: The biggest tech news coming out of China WIRED may earn a portion of sales from products that are purchased through our site as part of our Affiliate Partnerships with retailers. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of Condé Nast.

On Monday, the New York Stock Exchange (NYSE) announced the development of its tokenized securities platform for trading U.S. stocks via blockchain technology. The new platform is said to enable trading on a 24/7 basis and immediate settlement of assets, among other features, and it will operate as a completely separate addition to the traditional NYSE. Notably, regulatory clarity around tokenized stock trading is a key component of the CLARITY Act, which is currently under discussion in Congress and recently experienced a setback in the form of crypto exchange giant Coinbase's removal of support. Despite Wall Street's increased interest in the concept of tokenization oftentimes being touted as a win for crypto platforms, CoinDesk has reported the blockchains being used by NYSE are private and not public platforms like Ethereum or Solana. Unlike the decentralized crypto networks that are more frequently reported on in the press, a private blockchain is a permissioned network where access and participation are restricted to authorized entities, often controlled by a single organization or consortium of banks. For example, stablecoin issuers Circle and Tether are now promoting their own blockchain offerings that have stablecoin-centric features and avoid costs associated with decentralization that may not be necessary for centrally-issued, dollar-pegged assets like USDC and USDT. That said, many private blockchain networks are compatible with the Ethereum Virtual Machine (EVM), which is a point often used by Ethereum proponents to show their technology is still having an impact, despite these private systems not having any connection to the public Ethereum network or its associated ETH crypto asset. Last year, Robinhood also outlined stock tokenization as the key value proposition of their own Ethereum layer-two network. This is also a major area of emphasis for more crypto-focused fintechs such as Coinbase and Kraken. At this point, it's unclear where tokenized stocks will ultimately end up trading, but what is clear is these sorts of centralized institutions built on top of crypto's base layers will want to maximize revenue and control, as has already been seen with the stablecoin issuers. Increased centralization around real world asset tokens (namely stablecoins) has been a key area of concern and contentious debates over the past year, as more tech giants and banks are coming to the space to issue stablecoins rather than simply become another node on a network like Bitcoin. Indeed, NYSE's announcement from Monday morning also points to the use of stablecoins as a funding mechanism. As crypto has continued to rely upon stablecoins and other centralized power structures in search of greater adoption, the sector moves further away from Bitcoin creator Satoshi Nakamoto's original design, which ironically focused on disintermediating those same sorts of financial and tech institutions. The Bitcoin network's focus on its own native asset (rather than centrally-issued tokens) has somewhat shielded itself from this trend towards centralization. However, growing institutional support as a long-term reserve asset, whether it be via the Harvard University endowment or the U.S. government, has led to growing concerns around the centralization of bitcoin custody. Arizona, New Hampshire, and Texas have enacted laws aimed at creating their own reserves. Turns out people might like to circumvent centralized financial infrastructure in times of political upheaval.