Silicon Valley Splits Over Chinese Open-Source AI
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Anthropic and OpenAI have lobbied Washington regulators to restrict Chinese open-source AI models, with Treasury Secretary Scott Bessent and Trump's science adviser Michael Kratsios weighing in on American AI models as valuable intellectual property.
- Nvidia CEO Jensen Huang made his first-ever post on X stating "the world needs both frontier closed models and frontier open models," followed minutes later by Microsoft CEO Satya Nadella declaring open-source software "essential to a healthy A.I. ecosystem."
- Microsoft, Nvidia, Meta, Palantir, and IBM executives signed a letter supporting open-source AI, arguing it advances the technology through shared knowledge — while Microsoft and Nvidia note that open models also drive demand for their cloud computing services and chips.
- Chinese startups Z.ai and Moonshot AI recently released models rivaling those from American labs, intensifying the regulatory debate.
- Anthropic and OpenAI allege Chinese companies built their models by improperly harvesting data from American AI systems and argue this practice should not be allowed.
Why it matters: If Anthropic and OpenAI succeed in restricting Chinese open-source models, Microsoft and Nvidia lose a key driver of cloud and chip demand, while Silicon Valley startups lose access to freely available competing models — potentially cementing the lead of the two firms lobbying hardest for the restrictions.
