VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

Major AI Firms Push Back Against Broad U.S. Restrictions on Open-Weight Models

Hugging Face, Meta, Microsoft, Mistral, and Nvidia signed an open letter opposing broad restrictions on open-weight AI as Washington weighs a response to…

Vector Wire — AI-assisted editorial illustration

Several major AI companies — including Hugging Face, Meta, Microsoft, Mistral, and Nvidia — have signed an open letter urging U.S. policymakers not to impose broad "premature restrictions" on open-weight AI models2.

The letter arrives as Washington debates how to respond to allegations that Chinese AI labs have been stealing intellectual property from American counterparts. The Trump administration has reportedly been considering banning Chinese open-weight models and potentially issuing sanctions against AI companies from China. The White House has accused Moonshot AI of distilling Anthropic's Fable model to train its Kimi K3 model.

The open letter does not mention China directly, but it appears aimed at discouraging a total ban on Chinese models and at ensuring the administration's response to alleged distillation does not spill over into broader restrictions on open-weight AI or common development techniques.

"Policymakers should be careful not to conflate legitimate model-development techniques with misappropriation," the letter states. It describes distillation — the practice of using one model's outputs to help train or improve another — as "a widely used technique for model improvement, evaluation, and validation" that "reflects a long tradition of learning from, building upon, and improving existing technologies".

Startup founders have separately urged the U.S. government not to shut off access to Chinese open-weight AI, according to a Politico report that drew substantial attention on Hacker News, accumulating over 1,000 points and more than 800 comments1.

Media commentary has framed the situation as the American AI industry attempting to navigate both a competitive threat from China and potential regulatory complications from the U.S. government. One commentator noted that Chinese open-weight models are priced cheaply or offered for free, disrupting the economics that major U.S. AI companies have relied on. Another observed that OpenAI and Anthropic charge significantly more for their models by comparison.

The debate also touches on the pace of AI development. One speaker noted that experts consistently say they do not know where the technology will be in six months, underscoring the difficulty of crafting durable policy in a fast-moving field.

ANALYSIS The coalition behind the letter is notable for its breadth: it spans open-source-focused organizations like Hugging Face, chip manufacturers like Nvidia, and frontier-model developers like Meta and Microsoft, all of which have commercial stakes in the open-weight ecosystem. Their joint intervention signals that proposed restrictions could affect not just Chinese labs but the broader infrastructure of open-weight model development and distribution that these companies depend on.

The letter's careful distinction between distillation as a general technique and distillation as potential IP misappropriation reflects a core tension in the debate: the same methods that the White House has cited as evidence of Chinese misconduct are standard practice across the industry. Broad regulatory action targeting these techniques could constrain American developers alongside their Chinese counterparts.

CORRECTIONS: none for this article · this piece updates automatically as the story develops · corrections policy & trail →