Skip to content
NewsRegulation

Fifty companies back open-weight AI letter that Jensen Huang amplified with his first-ever X post

· by Pondero Newsdesk

The short version

A 50-company coalition published an open letter urging Washington not to restrict downloadable AI models. Nvidia CEO Jensen Huang debuted on X to amplify it. OpenAI joined the second wave. Amazon and Anthropic stayed out.

Fifty companies back open-weight AI letter that Jensen Huang amplified with his first-ever X post

OpenAI quietly signed a letter opposing AI model restrictions even though the company sells closed, proprietary frontier access. That detail is the most unexpected element in a 50-company coalition that, as of July 25, 2026, has become the highest-profile industry push against a potential US government ban on freely downloadable AI weights. Amazon and Anthropic stayed off the list entirely. Per Forbes, the count doubled from 25 to 50 in a single day after Nvidia CEO Jensen Huang used his first-ever X post to amplify the letter.

What happened

On July 24, 2026, Nvidia, Microsoft, Meta, IBM, Palantir, Mozilla, Mistral, Hugging Face, the Linux Foundation, Dell, and Andreessen Horowitz co-signed "Open Weights and American AI Leadership," hosted on Microsoft's On the Issues blog. The letter asks Washington not to impose restrictions on downloadable, modifiable AI models, warning specifically against "premature restrictions" it said would weaken smaller developers and cede ground to foreign competitors.

Jensen Huang joined X the same day and used his first post to share the letter, writing that "open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty," per HotHardware. His post cleared 11 million views, per Forbes. Huang had surpassed 70,000 followers on the platform within hours of the post.

By July 25, the signatory count had doubled to 50. OpenAI, Google, AMD, Cisco, Cloudflare, GitHub, Block, and Ollama were among the 25 companies that joined the original roster, per Forbes. The expansion happened while the news cycle around the original post was still running. No single coordinating announcement preceded the additions.

Why it matters

The letter is a direct lobbying document. If Washington moves to restrict open-weight models, every team running Llama, Mistral, or any other publicly downloadable weight faces a narrower set of options. The alternative is API access through proprietary providers, at per-token pricing, without the ability to fine-tune or self-host. That is a material cost increase and a loss of control for anyone who has built workflows around open weights.

OpenAI's participation stands out. The company built its business model on closed, subscription-gated access. Signing alongside Meta and Mistral, which distribute model weights publicly, places OpenAI on record in favor of the exact distribution model it does not use. No official statement from OpenAI accompanied the addition, per TechRadar.

The letter's core argument is that open architectures support safety research rather than undermining it. Per the letter, as quoted by HotHardware: "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." That framing is aimed directly at the policy debate around whether open-weight models give adversaries a capability advantage.

For operators choosing long-term AI vendors, the roster is a signal about whose interests align with open-weight access remaining legal. The 50 signatories have a stated position. Amazon and Anthropic do not.

Context and reactions

The letter's timing connects to US government scrutiny of Chinese open-weight models. The White House accused Moonshot AI, the Chinese firm behind the Kimi K3 open-weight model, of distilling output from Anthropic's AI systems to build its own competing model. Moonshot AI denied the allegation, per TechRadar. The administration has weighed trade restrictions targeting freely downloadable weights from Chinese firms, and the letter is a response to that specific policy direction.

Amazon is Anthropic's largest investor. Anthropic has argued publicly that AI model weights, once released, cannot be recalled and that this irreversibility makes open distribution inherently riskier than closed, access-controlled deployment. Google is also an Anthropic investor but signed the letter, which sharpens the Amazon-Anthropic position as a distinct stance rather than a general investor instinct toward caution.

Forbes noted a structural tension in the coalition: Nvidia organized the letter while owning CUDA, the proprietary software stack that makes its hardware genuinely difficult to replace. The letter advocates for openness at the model layer, which is where none of the signatories hold their primary competitive position. Each signatory is, in effect, arguing for freedom in a layer they do not own.

The speed of the count doubling, from 25 to 50 in under 24 hours after Huang's post, reflects how policy coalition formation now works in AI. A visible distribution event, not a coordinated sign-on process, drove the expansion. Companies joined while coverage was still in progress.

What to watch next

Whether Amazon and Anthropic publish a formal counter-statement or join a later signatory wave will indicate how far apart the industry's two lobbying positions actually are. Commerce Department and White House rulemaking on open-weight Chinese models is ongoing; the letter's framing will either find traction with the administration's stated goals around sovereignty and competitiveness, or the IP-theft narrative around Moonshot AI will drive a harder line. Teams running open-weight models in regulated industries should track both the policy language and whether any formal restrictions emerge targeting specific model origins.

Sources