AI工具Score A (69)
Jensen Huang's Nvidia Ouster Removes Major Threat to Anthropic IPO - Business Insider
5 小时前2 viewsSource: businessinsider.com
Fortune/NurPhoto Pictures via Reuters Connect Nvidia cofounder Jensen Huang, left, and Anthropic CEO Dario Amodei. Listen 5m 2026-11-17T10:00:02.437Z Save Saved Read in app This story is available exclusively to Business Insider subscribers. Become an Insider and start reading now. Have an account? Log in . Jensen Huang 's dramatic departure from Nvidia may be the best thing to happen to Anthropic just as the AI startup prepares to go public this week. Under Huang, Nvidia was investing heavily in cheaper, more open AI models that compete with Anthropic's offerings. The chip giant was also backing startups developing these alternative models and helping make them widely available. That threatened Anthropic's business model, which involves spending heavily to develop closed AI models and charges customers to use them, often at higher prices than open alternatives. "People really need to understand that the 'Cold War' going on here is that Nvidia was on a titanic collision course with the labs," the CEO of a leading AI startup told Business Insider. This person asked not to be identified while discussing sensitive topics. The executive, who supports the open-source AI community and is sympathetic to Huang, described an ambitious Nvidia strategy to develop its own AI models, invest in rival startups, and build an ecosystem that could challenge not just Anthropic, but OpenAI too. This strategy was just beginning to come together. But with Huang gone, the threat has diminished considerably. The timing is extraordinary. Anthropic is preparing for a stock market debut this week that could value it at $1.4 trillion. Investors are being asked to make an enormous bet on the future profitability of its AI services. Huang was helping build a world in which companies wouldn't necessarily have to pay Anthropic for that intelligence. Nvidia's open AI push Amjad Masad, chief executive officer of Replit David Paul Morris/Bloomberg via Getty Images The strategy centered on " open-weight " AI models. Unlike Anthropic's Fable and Opus offerings and OpenAI's top GPT models, these alternative systems can be downloaded, modified, and operated independently. Want more Business Insider in your news feed? Add BI in Google so our reporting is easier to find when you’re searching for what matters. Add to Preferred Sources That makes them attractive to businesses that want greater control over their AI systems without paying a leading AI lab every time someone uses the technology. Nvidia was pursuing this aggressively. It developed its own family of open models, called Nemotron. It also backed startups including Reflection AI, Thinking Machines Lab, and Poolside, while moving to acquire Hugging Face, a major platform for sharing open-weight AI models. Replit CEO Amjad Masad called Jensen Huang's ouster "an attack on open models" at a Silicon Valley conference on Monday. "Jensen is the premier champion of open models," Masad said. "If it wasn't for Jensen, there wouldn't be a lot of funding going into US open model companies." Fireworks Fireworks AI CEO Lin Qiao (left) interviewing Jensen Huang at the startup's conference in San Francisco on Nov. 3, 2026. Alistair Barr/Business Insider On November 3, I watched Huang evangelize this strategy at a Fireworks AI conference in San Francisco. He acknowledged the value of closed AI models from companies such as Anthropic and OpenAI, but argued that businesses shouldn't have to rely on them. Instead, they could take open-weight models and adapt them to their specific needs, creating AI systems that could be even more powerful. "By taking an open frontier model, adapting it and providing it with your domain knowledge, your access, your specialized tools, and the proper evaluation, you should be able to create essentially a super intelligence," Huang told Fireworks CEO Lin Qiao on stage. "And I think those times are here," he added. Nvidia's worst nightmare This strategy made business sense for Nvidia. More open models running on its chips meant more potential customers. It was becoming increasingly important as Anthropic and OpenAI sought alternatives to Nvidia's expensive hardware. In fact, Anthropic has developed a system that could be Nvidia's worst nightmare, as Business Insider reported in October . Anthropic disclosed in its IPO filing that it has spent years developing software that allows it to shift AI work between different types of chips from Nvidia, Google, Amazon, and AMD. That gives Anthropic more flexibility to choose the cheapest or most effective chips for a particular task, reducing its dependence on Nvidia and potentially weakening Nvidia's ability to charge premium prices. The shift is already significant. A November 14 analysis by Michael Cembalest of JPMorgan Asset Management estimated that Nvidia accounted for just 39% of future computing commitments by leading AI labs, down from more than 80% of AI chip revenue in recent years. The China battle There was another key area of AI where Nvidia and Anthropic clashed. Anthropic CEO Dario Amodei has advocated for very strict limits on advanced US chip sales to China, arguing these restrictions are essential to maintaining America's technological lead. Huang fought against those restrictions, warning that shutting China out would encourage Chinese companies to develop competing technology. Now he's gone, Anthropic's worldview is prevailing. On Monday, the US government rolled out new restrictions that close a loophole that allowed Chinese firms to use Nvidia GPUs by renting them from data centers outside China. It's a win for China hawks within the administration and a huge victory for Amodei, arguably the biggest China hawk of them all. There are caveats. For instance, Nvidia's interim CEO could continue Huang's strategies. Still, the immediate strategic implications are striking. Anthropic is about to ask investors to bet on its future dominance of AI. Just days before that moment, one of the industry's most powerful figures pushing an alternative vision has been forced out. For Anthropic, that's enormous. Sign up for the AI Insider newsletter here . Reach out to me via email at abarr@businessinsider.com
Read the full original article:
businessinsider.com