More than 20 companies signed a letter warning that premature limits on open-weight AI could stifle competition and push innovation overseas
Anadolu / Getty Images
A coalition of more than 20 technology companies, including Nvidia $NVDA, Meta $META, Microsoft $MSFT, and Palantir $PLTR, released a joint letter on Friday urging policymakers to avoid placing restrictions on open-weight artificial intelligence models.
Open-weight AI models are systems whose weights are publicly released, allowing anyone to deploy, examine, and adapt them on hardware they control. The letter argues that such models expand access to the AI economy, strengthen competition, and give organizations greater control over the technology they use.
"Our AI leadership will be judged not by one frontier AI model, but by whether the United States builds a strong, open ecosystem that diffuses into every sector," the letter said.
The signatories warned that relying only on closed models carries its own dangers. "Relying solely on closed models is not inherently safe: they can be breached, misused, or fail in ways that outsiders cannot detect," the letter said. "And concentrating advanced AI capabilities behind a small number of closed models compounds that risk."
The letter also addressed concerns about AI distillation — a technique in which a smaller model is built using outputs from an existing, stronger model. The companies called for any concerns about unlawful distillation to be handled through "targeted legal and commercial frameworks" rather than restrictions on the technique broadly.
Nvidia CEO Jensen Huang and Microsoft CEO Satya Nadella both shared the letter on their personal social media accounts, according to CNBC.
The letter comes amid a wider debate over Chinese open-weight AI models. Moonshot AI, a Chinese startup, sharpened those concerns earlier this month when it unveiled a model called Kimi K3 that has beaten several U.S. competitors on various industry benchmarks. U.S. Treasury Secretary Scott Bessent said on Tuesday that the Trump administration would examine whether Chinese companies were stealing American intellectual property.
White House advisor Michael Kratsios alleged on Wednesday that Kimi K3 was built by distilling outputs from Anthropic's models, according to CNBC. Kratsios argued that while the technique has legitimate uses that support open innovation, deploying it covertly at scale to misappropriate American technology crosses a line he called "unacceptable."
OpenAI and Anthropic did not sign the letter. OpenAI president Greg Brockman said Thursday that he had not taken part in any White House discussions about restricting access to Chinese open-weight models. "I think that, fundamentally, AI and AI usage is something that is actually very important to democratize," Brockman said.
Join 500,000+ readers who start their day with Quartz.
By subscribing, you agree to our Terms of Service and Privacy Policy.
