More than 25 technology companies signed an open letter on Friday urging US policymakers not to impose broad restrictions on open-weight AI models, landing while the Trump administration weighs how to respond to Chinese open-model releases.
Who signed
Confirmed signatories include Hugging Face, Meta, Microsoft, Mistral, Nvidia, Replit and a16z. The full roster runs past 25 names; individual outlets have published only partial lists, so a six-name summary is an excerpt rather than the whole.
Who did not
OpenAI, Anthropic, Google DeepMind and SpaceX are confirmed absent. That absence is the story: the frontier labs with the most valuable closed weights sat out a letter defending open ones.
The argument
The letter defends distillation as a widely used technique for model improvement, evaluation and validation — directly relevant, because the trigger was the White House accusation that Moonshot AI distilled an Anthropic model to build Kimi K3. Signatories warn that premature restrictions would stifle competition or push innovation overseas.
Why Microsoft's signature is the odd one
Microsoft sells closed frontier models through Azure and its own MAI line, so its interest in open weights is not ideological — open models are commodity inference demand, and commodity demand is what a hyperscaler wants. Meta's position is more direct: Llama is the open-weight family a US restriction would damage most. Nvidia's is simpler still, since every open model that runs anywhere runs on its silicon.
Nothing has been proposed yet
There is no bill, no rule and no agency deadline attached to this. The administration is weighing a response; the letter is pre-emptive lobbying against a policy that does not formally exist. Jensen Huang marked the occasion by making his first post on X to endorse it.