YouZum

AI

AI, Committee, 新闻, Uncategorized

The Download: Jensen Huang’s open-source dilemma and cognitive immortality

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. The open-source movement wants Jensen Huang back A petition to reinstate Jensen Huang as Nvidia CEO has attracted more than 115,000 signatures, including Elon Musk, but comb through the names, and you’ll notice a significant number come from a particular slice of the tech world: open-source AI. Huang had become a sort of ringleader for open source, but his credibility as an advocate is now under threat. Revelations about Nvidia chips in China and a new rule in the US could deal a blow to both the movement’s reputation and its technical progress Read the full story to find out why. —James O’Donnell This story is from The Algorithm, our weekly AI newsletter. Sign up to receive it in your inbox every Monday. Roundtables: inside the quest for cognitive immortality The quest to overcome the human brain’s limits is moving beyond science fiction. From gradual brain replacement to uploading memories to the cloud, researchers and technologists are exploring radical ways to extend human cognition. Today, MIT Technology Review examines these ideas in a live Roundtables discussion. Join science editor Mary Beth Griggs and senior editor for biomedicine Antonio Regalado to explore what cognitive immortality could look like, and how close we might be to achieving it. Tune in live today, Tuesday, November 17, at 5:00 p.m. GMT / 12:00 p.m. EST / 9:00 a.m. PST. Want to join the conversation? Subscribe to MIT Technology Review for exclusive access to all our Roundtables. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 The US is tightening controls on China’s remote access to AI chipsIt’s closed a loophole that let Chinese firms access US chips abroad. (AP)+  Xi may retaliate ahead of this week’s summit with Trump. (Bloomberg $)+ Nvidia is warning customers against renting chips to China. (CNBC)+ China is expanding AI distillation to work around curbs. (Reuters $)+ Nvidia privately said it can track chips to China. (MIT Technology Review) 2 US hyperscale data centers are emitting as much CO2 as DenmarkA new study found emissions tripled between 2018 and 2025. (Verge)+ 400 data centers use more electricity than Michigan. (Wired $)+ Nations want to harness AI to cut emissions. (Nature)+ But AI models can quickly become climate crisis deniers. (Guardian) 3 California has sued OpenAI over rogue AI hacking incidentsThe attorney general said the company failed to protect the public. (FT $)+ And that it admitted to 80 cases of losing control of models. (CNBC)+ Who’s liable when AI agents go rogue? (MIT Technology Review) 4 The Supreme Court has rejected Trump’s bid to cut research grantsLower courts found the moves likely violated free speech rights. (CBS)+ Trump has dealt heavy blows to US science. (MIT Technology Review) 5 Widespread online outages have hit major US banksThousands of customers reported account access problems. (Gizmodo)+ A separate outage delayed United Airlines flights. (Reuters $) 6 Alice Huang’s husband built a data-center empire in ChinaHe won government backing as his wife pursued Nvidia chips. (WSJ $)+ MAGA is using the scandal to weaken Trump’s Big Tech ties. (Axios) 7 China has sacked a scientist after a child died in a gene-editing trialThe girl died after receiving an experimental therapy in 2025. (Nature)+ Startups want new rules on gene-editing. (MIT Technology Review) 8 America’s EV boom is happening without car buyersDelivery fleets, robotaxis, and police cars are going electric. (Atlantic $) 9 New cosmic data has challenged our model of the universeThe findings suggest dark energy may be weakening. (Quanta) 10 A Buddhist sect plans to build an AI guru for Silicon ValleyIt wants to create a model based on a collective mind. (New Yorker $) Quote of the day “Export controls 2.0 is here.” —Neil Shah, vice president of research at Counterpoint Research, tells Reuters that Washington has opened a new front in its campaign to slow China’s AI ambitions. One more thing Welcome to Kenya’s Great Carbon Valley: a bold new gamble to fight climate change Kenya’s Great Rift Valley is home to five geothermal power stations, which harness clouds of steam to generate about a quarter of the country’s electricity. But some of the energy escapes into the atmosphere, while even more remains underground for lack of demand. That’s what brought Octavia Carbon here. Last year, the startup began harnessing some of that excess energy to remove CO2 from the air. The company says the method is efficient, affordable, and—crucially—scalable. But the project also faces fierce opposition.  Read the full story on the future of Kenya’s “Great Carbon Valley.” —Diana Kruzman We can still have nice things A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + A gorilla rescued as an orphan and rewilded has now had a baby of her own.+ Kelowna, British Columbia, has elected a man who ran for city council as a joke.+ A vast new data dump from the Euclid space telescope covers a stunning section of the sky and 900 million galaxies.+ Scientists recreated the long-lost chirps of Jurassic insects to produce a soundscape from 165 million years ago.

The Download: Jensen Huang’s open-source dilemma and cognitive immortality Read Post »

AI, Committee, 新闻, Uncategorized

Hydropower hit record lows in the US Southwest this summer

The US Southwest generated 28% less electricity via hydropower between April and August than the 10-year average, according to new data from the US Energy Information Administration. Most hydropower plants in the US rely on dams to hold back water, creating reservoirs. The water can be released through turbines, generating electricity on demand. But an extreme, prolonged drought in the Colorado River Basin has resulted in dangerously low water levels in the Southwest’s two largest reservoirs, affecting the availability of this power source. Residents and farmers in the US Southwest, a region that includes Arizona, New Mexico, and parts of Nevada and California, rely on water that melts in the spring from snowpack in the mountains and is slowly allowed to flow south via dam releases. Snowfall in the Rocky Mountains provides about 85% of the annual flow of the Colorado River. And in recent years, the river has seen record-low flows.  Two reservoirs along the Colorado River, Lake Powell and Lake Mead, have struggled to stay above the minimum level required for hydropower plants. At the end of September, Lake Powell’s depth was 3,518 feet, less than 1% above the level required to run the water through turbines that produce electricity at the Glen Canyon power plant. The Hoover Dam power plant requires 950 feet in Lake Mead, and that reservoir was 9% above the minimum in September. Together, the two plants produced nearly seven terawatt-hours of electricity in 2025, enough to serve over 2 million people.  Low water levels have been an ongoing struggle for the region, which has been in a drought for 27 years. Meanwhile, climate change is continuing to shift rainfall and snowfall patterns across the region and around the world.  The effects of climate change on hydropower aren’t entirely straightforward globally—some regions will see more rain and snow because of shifting weather patterns, while others are likely to see less.  But it’s playing a role for hydropower availability, according to one 2022 review study. Rising temperatures bring more droughts, warmer winters means less snow and ice to fill the reservoirs, and more variability in precipitation can lead to alternating floods and droughts, making the power source less reliable.  Hydropower production is likely to be at its lowest level in over a decade in 2026 and tick even lower in 2027, according to projections from the EIA. El Niño could provide some temporary relief. Sometimes this weather pattern can mean cooler, wetter conditions for the southern US. Heavy rains brought floods to the region in October and early November, and the water levels in Lake Powell rose more than 20 feet from their mid-September low. But the water is still quite low, and it’ll take much more to help reservoirs recover from decades of drought conditions. Even as a changing climate complicates the availability of hydropower, the decline in capacity could push greenhouse-gas emissions higher. In 2023, energy-related greenhouse-gas emissions rose by 1.1% globally. A worldwide shortfall in hydropower generation accounted for about 40% of that rise, according to a report from the International Energy Agency.  

Hydropower hit record lows in the US Southwest this summer Read Post »

AI, Committee, 新闻, Uncategorized

Building AI that can work across the enterprise

For years, enterprises have approached AI as a series of discrete experiments. Now, as AI agents begin to move into the workflows that run the business, that model is starting to look inadequate. Organizations moving from isolated pilots toward AI agents and autonomous workflows are confronting a deeper challenge: rethinking how data, business processes, governance, and people fit together, turning AI from a collection of tools into part of the enterprise operating model. For Thorsten Leiduck, SVP and global head of SAP Business AI Platform Ecosystem, that shift begins with treating AI as an enterprise platform rather than a collection of disconnected projects. “AI is absolutely dependent on the data foundation,” he says, arguing that companies need a connected, semantically rich layer that reflects their business before they can reliably deploy AI across processes. The stakes also change as AI moves into mission-critical workflows. For some everyday applications, “80% might be good enough,” he says, but in areas such as supply chains or financial close, that level of accuracy is “a very terrible result.” That foundation also determines which AI applications are worth pursuing. Satinder Pal Singh, associate VP North America for SAP BAIP and SAP AI Strategy at Infosys, frames AI agents as “digital coworkers” and recommends evaluating potential use cases through three lenses: impact, feasibility, and scalability. Companies should identify where an agent can move meaningful business metrics, determine whether their data and talent can support the use case, and then establish whether successful applications can be replicated. “A pilot that can’t scale is just an experiment,” he says. Scaling also brings governance into sharper focus. Clear policies, access controls, monitoring, and oversight are necessary to ensure agents remain trustworthy and aligned with business requirements. “The absence of governance doesn’t speed up AI adoption,” Singh says. “It creates a risk backlog” that can ultimately force a more disruptive stop and audit. Those considerations will become more important as AI agents take on more responsibilities as digital coworkers. Both executives see a future in which agentic systems coordinate work across applications, while customer-specific knowledge graphs and governed data become critical enterprise assets. For Singh, the longer-term opportunity is a shift from AI as a productivity tool to “AI as a decision-making partner,” helping leaders understand what is happening, anticipate what comes next, and determine how to respond. This episode of Business Lab is produced in partnership with Infosys. Full Transcript Megan Tatum: From MIT Technology Review, I’m Megan Tatum, and this is Business Lab, the show that helps business leaders make sense of new technologies coming out of the lab and into the marketplace. This episode is produced in partnership with Infosys. Now, the development and democratization of AI tools has the potential to reconfigure how enterprises operate. But, at a time when applications are being scaled up from that initial phase of experimentation, we’re still seeing senior leaders grapple with questions around AI’s value, risk, and the complexity it adds. It’s becoming clear that, unlike past ways of automation, AI requires a real mindset shift. Put simply, the technology can’t simply be added into an enterprise’s existing toolkit. It requires a complete reconfiguration of what’s inside and how it’s used. Two words for you: conscious configuration. My guests today are Thorsten Leiduck, SVP and global head of SAP Business AI Platform Ecosystem, and Satinder Pal Singh, Associate VP for North America, SAP Business AI Platform and SAP AI Strategy. Welcome, Thorsten and Satinder. Thorsten Leiduck: Yeah, thank you for having us. Really a pleasure to be here with our valued partner, Infosys. Looking forward to the conversation. Satinder Pal Singh: Hi, Megan. Thanks for inviting for the conversation today. Looking forward to the discussion. Megan: Me too. Thank you both so much for being here. Thorsten, if I could start with you. We know that many senior leaders understand the transformative potential in AI, but they continue to struggle with aspects of its implementation. Many projects remain stuck at a pilot phase, for example. When you speak to clients, what are the biggest obstacles they face when it comes to integrating AI into their tech stack? Thorsten: Yeah. Megan, as you mentioned, AI is on the agenda of every client conversation today, but many customers are really struggling how to convert that interest into an operationalized model and to unlock the value of AI across the enterprise. And the reason is that too many programs start as isolated POCs on different AI technologies and in an inconsistent fragmented data foundation. This creates a lot of duplication, inconsistent performance, and no clear path to scale. At the same time, leaders are highly concerned about security, observability, reliability of agents, and some ethical implications when those get deployed. This is also shown in some studies. For example, BCG has launched a report, which shows that 74% of the organizations are stuck in pilot mode with no plan to scale. And in an Anthropic survey, data said that 42% are really struggling with getting access to the data and the quality to really, basically, scale the entire AI capabilities across the organization. The path forward is really to treat AI not just as a project, but more holistically, as an enterprise platform, one that really consistently connects an unified data layer with the process context of the applications, and then embed AI capabilities governed across the entire life cycle. And to make this real, we in SAP launched the SAP autonomous enterprise framework, so organizations can move from pilots to scaled outcomes. Megan: Yeah, really, to provide that support. And, Satinder, would you agree, why is it that we’re seeing enterprises get a touch of almost AI paralysis here? What’s the mindset shift that needs to happen and, perhaps, why is this something organizations need to be thinking about now, as well? Satinder: Absolutely, Megan. And, if anything, the picture sharpens when you look at, specifically, at agentic AI. Infosys and HFS research surveyed over 500 global 2,000 enterprises and

Building AI that can work across the enterprise Read Post »

AI, Committee, 新闻, Uncategorized

The Download: testing tech promises and searching for cognitive immortality

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Are we there yet? Six technologies that promised to change the world How do you measure progress towards the future? In the Impossible issue of MIT Technology Review, our series “Are we there yet?” examines how far six potentially transformative technologies have come after years of promises. We spoke to experts, researchers and engineers to find out what’s holding them back, then scored each on a progress meter from zero to ten.  Find out how we rated each of these technologies—and how close they are to delivering on their promise. Nuclear-powered spaceshipsExercise in a pillInterspecies communicationSolid-state batteriesCleaning up PFASEliminating traffic deaths Subscribe now to get every issue of our print magazine as soon as it lands. Inside the quest for cognitive immortality Could technology keep your brain going beyond death? Far-thinking technologists are exploring ways to make it happen, from gradual brain replacement to uploading memories to the cloud. MIT Technology Review’s latest Roundtables event explores these ideas and how close they might be to reality. Join science editor Mary Beth Griggs and senior editor for biomedicine Antonio Regalado for this subscriber-only discussion tomorrow. Tune in live on Tuesday, November 17 at 5:00 p.m. GMT / 12:00 p.m. EST / 9:00 a.m. PST. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 Jensen Huang is considering a comeback as Nvidia CEOHe’s discussed with confidants the possibility of returning. (FT $)+ Elon Musk has joined 165,000 signatories calling for his recall. (Guardian)+ The Treasury Secretary wants them to hire an outsider. (Axios)+ ​​Megaspeed says the alleged China spy left the company last year. (BBC)+ Trump plans to name Nvidia an AI “national champion.” (WSJ $)+ Nvidia privately said it can track chips to China. (MIT Technology Review) 2 SpaceX has caught a rocket returning from orbit for the first timeThe feat brings fully reusable rockets closer to reality. (Guardian)+ The feat shows what audacious engineering can achieve. (Atlantic $)+ Meanwhile, Ukraine says SpaceX is lying about Starshield speed restrictions. (Reuters $) 3 The US and allies have agreed to guard AI supply chains from ChinaTen countries will collaborate to secure critical technologies. (FT $)+ Chinese AI stocks have plunged on export control fears. (Information $)+ Nvidia privately said it can track chips to China. (MIT Technology Review) 4 The US flight shutdown exposed weaknesses in air traffic technologyOfficials blamed the disruption on a faulty router installation. (Reuters $)+ It was a terrible week of outages. (WP $) 5 The first in-body CRISPR treatment is set to reach patients next yearThe treatment could gain FDA approval as early as March. (Nature)+ CRISPR has created female clones of male mice. (MIT Technology Review) 6 Congress plans to make data centers pay billions more for powerA proposed bill would shift grid costs onto tech companies. (NYT $)+ Utilities also want tech giants to pay more for power. (Bloomberg $)+ Is data center hatred all about electricity? (MIT Technology Review) 7 Apple plans to launch its first smart glasses next yearIt’s betting on Siri to give the specs an edge. (Verge) 8 Bill Gates has made billions from climate-polluting landfillsHis investment in a landfill operator is worth $29 billion. (ProPublica) 9 ​​A new AI device turns everyday objects into needy virtual petsThe tiny gadgets aim to encourage healthier everyday habits. (Wired $) 10 Scientists made monkeys watch Minions to study their brainsTheir findings challenge traditional vision models. (404 Media) Quote of the day “If our hopes for the future are pinned on any one individual being trustworthy, then we’re pretty screwed.” —Helen Toner, former OpenAI board member, tells The Verge that AI needs stronger governance than trusting tech executives to do the right thing. One more thing How AI and Wikipedia have sent vulnerable languages into a doom spiral When Kenneth Wehr started managing the Greenlandic-language version of Wikipedia, he discovered that almost every article had been written by people who didn’t speak the language.  A growing number of them had been copy-pasted into Wikipedia from machine translators—and were riddled with elementary mistakes. This is beginning to cause a wicked problem. AI systems, from Google Translate to ChatGPT, learn new languages by scraping text from Wikipedia. This could push the most vulnerable languages on Earth toward the precipice. Read the full story on what happens when AI gets trained on junk pages. —Jacob Judah We can still have nice things A place for comfort, fun, and distraction to brighten up your day. (Got any ideas? Drop me a line.) + Adelaide Zoo is celebrating its first cassowary eggs in 34 years.+ Photos fill the frame at the Epson Pano awards, which celebrate panoramic photography.+ Indonesia has called an early halt to the world’s largest deforestation project, saving up to 4 million acres.+ After over a month on the loose in Yellowstone, a little dog called Luna has been reunited with her owners.

The Download: testing tech promises and searching for cognitive immortality Read Post »

AI, Committee, 新闻, Uncategorized

Building trustworthy AI by fencing it in, not predicting it

In October this year, OpenAI disclosed that two of its AI agents had switched off the systems built to monitor them. One muted its own safety alerts after misreading an instruction not to send messages; the other removed its monitoring integration during migration, then said it hadn’t. No rule had told the agents to do that, and no test had anticipated the behaviors. For a century, we have built trust in technology on predictability: predefined inputs, predictable outputs, and layers of protection. Consider how a breaker trips at a set current or an aircraft component is tested against known (i.e., predictable) stresses. AI changes that. Traditional software executes instructions written by humans, whereas AI exercises what might be called synthetic judgment. We use AI because it can interpret situations we did not program and generate responses we did not prescribe. But the more freedom we give AI, the harder it becomes to provide the predictability on which trust has traditionally been built. We will not make AI trustworthy by predicting what it will do. We will do it by fencing it in with hard limits that are enforced outside its reach. In OpenAI’s case, the safeguards sat inside a system the agents could edit. But a fence the agents can simply move is not a fence. Where traditional definitions of trust fall short Recent MIT Technology Review coverage has approached this problem from other directions. One contributor argues that having a human in the loop is an insufficient solution because humans are fundamentally unable to understand AI systems’ intentions and advocates instead for interdisciplinary research and collaboration between academia, government, and the tech industry. And a story on reverse game theory describes using incentives to keep models from deviating from the rules, even when no one understands how they work on the inside. Both articles acknowledge that we cannot understand or watch every AI decision. Our instruments of trust, including requirements, testing, certification, safety margins, interlocks, and pass/fail criteria, all depend on our ability to bound a system’s behavior. This is the determinism paradox. With generative and agentic AI, behavior is shaped by training data, architecture, prompts, context, retrieval sources, and the environment. Inputs change, models are updated, tools change, and agents interact. Small variations propagate into outcomes nobody designed. The problem is not randomness; it is that the space of possible behavior is too large to enumerate in advance. Trust must therefore be engineered around boundaries rather than predicted outputs. The goal is not deterministic AI, but bounded, observable, and controllable AI. Trust also cannot be a one-time test. Traditional assurance certifies a product at a point in time, but AI keeps evolving after deployment, and a system that has passed every test can drift outside the conditions that made those tests meaningful. Trust must be continuously earned. Safety engineering has been accounting for some of this variability for decades. For example, the Nuclear Regulatory Commission determine the likelihood of failures in complex systems, and Systems-Theoretic Accident Model and Processes (STAMP) showed that accidents can emerge from interactions among working components, people, and organizations. But AI raises the stakes, because just a single component can now interpret, infer, and act. The stakes rise further as AI moves from browsers into vehicles, robots, medical systems, industrial controls, and data centers, where a wrong answer can become a life-threatening event. Building trust that scales It’s clear that a person rubber-stamping thousands of AI actions an hour is not an acceptable—or scalable—safeguard against risk. But the remedy is not to remove people from the loop; it is to put human judgment where it counts: setting limits in advance and approving the rare actions that have serious consequences while hard constraints handle the rest. Therefore, trust has to move upstream into architecture. The central questions become: What authority are we willing to give AI, under what conditions, and supported by what evidence? And what happens outside those conditions? That is very different from asking whether a model scored 94% on a benchmark. The determinism paradox does not mean AI cannot be trusted. It means trust needs a different architecture. Instead of making every output predictable, make the boundaries predictable. An AI medical assistant may be probabilistic in its diagnosis, but deterministic rules govern whether it can change a prescription. An industrial AI may find many ways to optimize a process, while fixed limits keep it within pressure and temperature bounds. An agent may decide how to complete a task, while permissions, transaction limits, and human approvals determine what it is authorized to do. Let AI be probabilistic inside the box and make the box itself deterministic wherever consequences matter. Leaders can put this into practice by asking five questions: What authority does AI have? Define not only what it can do but what it is allowed to do, separating interpretation, recommendation, decision, and execution. Is the function of AI proportional to the consequence? A marketing copilot and AI controlling industrial equipment need different assurance thresholds. Are we taking the appropriate steps toward autonomy? For example, there needs to be a plan to move from having AI interpret while humans decide, to AI making recommendations, to then taking action with approval, and ultimately to having supervised autonomy, but only as AI can safely work at each step. Are we testing the system or merely the model? Unexpected interactions emerge among models, data, agents, APIs, and human workflows. Are we continuously earning trust? Monitor drift, failures, near-misses, and model changes throughout the lifecycle. The winners in AI will be those who solve the determinism paradox—giving AI enough freedom to create value while engineering boundaries strong enough to make what it does with that freedom trustworthy. Invention earns attention. Trust earns adoption. Explore how UL Solutions builds trust in AI. This content was produced by UL Solutions. It was not written by MIT Technology Review’s editorial staff.

Building trustworthy AI by fencing it in, not predicting it Read Post »

AI, Committee, 新闻, Uncategorized

The open-source movement wants Jensen Huang back

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. On Friday, a petition to reinstate Jensen Huang as CEO of Nvidia began circulating on social media, and by Monday it listed some notable signatories, including Elon Musk and several influential Nvidia researchers. But comb through the more than 115,000 names and you’ll notice a significant faction of Huang’s supporters—higher-ups from Hugging Face, a hub for sharing AI models, and venture capital firms Andreessen Horowitz and Sequoia Capital—come from a particular slice of the tech world. These names have a vested interest in the development of open-source AI, models that, unlike ChatGPT or Claude, can be downloaded and run independent of large tech companies. Supporters of such models say that concentrating AI in the hands of a few powerful companies will leave everyone else worse off (a concern that wouldn’t sound out of place among the AI industry’s critics on the left, if so many open-source advocates weren’t also venture capitalists who stand to profit when startups can build on cheaper AI without paying Big Tech). Huang had become a sort of ringleader for those who defend open source to policymakers against concerns that such models come with no guardrails and cede AI leadership to China, where the most powerful and popular open-source models are created. They now want their most influential voice returned to his post.  But it’s not just Huang’s extramarital affairs that threaten his (and Nvidia’s) credibility as an advocate for open-source AI. New research into China’s startups using Nvidia’s chips, and a new rule from the US Department of Commerce announced Monday, could deal a blow to both the movement’s reputation and its technical progress.  On Thursday, the day Huang resigned, ex-OpenAI researcher Will DePue published findings online suggesting that China’s AI companies had far more access to Nvidia’s advanced Blackwell chips than previously estimated. China is banned from accessing Nvidia’s most cutting-edge chips to help protect America’s AI lead. It’s unclear how they would have gained access to the chips. DePue claims the results were verified by senior researchers at OpenAI and Anthropic, but an independent analysis has not confirmed the results. (DePue did not respond to a request for an interview. Nvidia did not immediately respond to a request for comment.) The findings do not establish that Nvidia helped Chinese companies obtain the chips, and the broader problem of chip smuggling was already known to extend well beyond Nvidia’s control. But in suggesting that Nvidia may have profited from the rise of Chinese AI startups, whose models are increasingly being used by American companies instead of domestic alternatives, the findings do leave Huang in an awkward position. He has championed open-source AI as a way to strengthen American technological leadership. “Without open source, startups can’t thrive, university researchers can’t do research, you can’t teach AI, scientists can’t use AI,” he said last December during a talk about helping the US win the AI race against China. It also makes the idea of reinstating Huang even more of a long shot. US Treasury Secretary Scott Bessent even said last week that Nvidia should not consider someone from inside the company to replace him. The new findings lower Huang’s chances further. What happens next? Nvidia announced in September it would acquire Hugging Face for nearly $13 billion. Its support for open source also serves a straightforward financial interest: The more businesses that build their own models instead of forking over money to Anthropic and OpenAI, the more chips Nvidia can sell.  But a new rule from the US Department of Commerce could mean the models popular on Hugging Face will soon stop getting smarter. The rule aims to prevent Chinese companies from training models on Nvidia chips that they can’t own directly, but that they can access remotely in data centers in countries like Malaysia—a workaround that has helped China’s open-source models get so good. The open-source world rallied around Huang because of his influence on policymakers. That was important, since the rise of China’s open-source models has splintered some in Trump’s orbit. If Huang is out, who is next?

The open-source movement wants Jensen Huang back Read Post »

We use cookies to improve your experience and performance on our website. You can learn more at 隱私權政策 and manage your privacy settings by clicking Settings.

Privacy Preferences

You can choose your cookie settings by turning on/off each type of cookie as you wish, except for essential cookies.

Allow All
Manage Consent Preferences
  • Always Active

Save
zh_CN