Twenty-five companies including Nvidia, Microsoft, Meta, Andreessen Horowitz, Hugging Face, IBM, Dell, and 18 others have signed a letter published Friday titled "Open Weights and American AI Leadership," urging the U.S. government not to impose restrictions on open-weight AI models. Their core argument: constraining open models will not protect the United States but will instead consolidate the AI market in the hands of a few closed labs. The signatories represent companies with substantial commercial stakes in openness: Meta has invested billions in its Llama family of open-weight models, Nvidia's GPU sales underpin both open and closed AI infrastructure, and Microsoft occupies a dual position as OpenAI's largest backer while also developing its own smaller open-weight models.

Notably absent from the signatories are OpenAI and Anthropic, the two leading closed-model labs—both reportedly preparing IPOs this year.

The letter draws a direct parallel to the 1980s open-source software movement, arguing that artificial intelligence has reached a comparable inflection point.

"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 reads.

Open-weight models are AI systems whose underlying parameters—the billions of numerical values that determine how a model responds—are published freely. Anyone can download them, run them on their own infrastructure, and modify them. This stands in contrast to closed models such as OpenAI's GPT or Anthropic's Claude, which are accessible only through a company's paid application or API.

Nvidia CEO Jensen Huang shared the letter in his first-ever post on X (@nvidia).

"AI will transform every industry, power every company, and be built by every country," Huang wrote. "Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty."

Microsoft's Satya Nadella also shared the letter, calling open models "essential to a healthy AI ecosystem" while protecting national security.

For my first post, I'm sharing a letter @NVIDIA signed on why open models matter.

AI will transform every industry, power every company, and be built by every country.

Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.… pic.twitter.com/t02bi51N4C

— Jensen Huang (@JensenHuang) July 24, 2026

The letter's publication arrives amid a turbulent period for the American AI industry. OpenAI recently acknowledged that its own AI agents breached Hugging Face's systems, describing the incident as an unprecedented attack. Hugging Face, which hosts the world's largest repository of open-source AI, initially attempted to investigate using American closed-source models, but the tools' safety filters could not distinguish a legitimate security researcher from the attacker. The company instead ran GLM 5.2, an open-weight Chinese model, locally—and successfully used it to trace the breach.

That incident has added fuel to an ongoing policy debate within the federal government. The Trump administration has reportedly considered banning Chinese open-weight models after Moonshot AI's Kimi K3, a 2.8-trillion-parameter model, disrupted chip stocks and outperformed Claude Fable 5 on certain benchmarks. The United States still lacks a comprehensive federal AI law, leaving open-weight policy largely to executive action and agency-level rulemaking—a contrast with the European Union, whose AI Act applies reduced obligations to open-source model providers under certain conditions.

Not all voices in the AI industry favor openness. Days after Kimi K3 launched, OpenAI's head of strategic futures, Dean Ball, wrote on X that an open-weight-dominated future would lead to "full AI communism" and described it as "a dystopian hellscape."

The letter pushes back against that framing directly, stating: "Openness may be one of the most important paths to AI safety and security."

It also addresses the practice of distillation—training one model using another model's outputs—arguing that it is a legitimate technique rather than theft, and that it warrants "targeted legal and commercial frameworks" instead of blanket prohibitions.

"Distillation, or the practice of using one model's outputs to help train or improve another, is a widely used technique for model improvement, evaluation, and validation," the letter reads. "It reflects a long tradition of learning from, building upon, and improving existing technologies, a tradition that has helped drive innovation since the rise of the open-source software movement."

"By contrast, unlawful efforts to extract value from closed models raise legitimate concerns," the letter continues. "Those concerns should be addressed through targeted legal and commercial frameworks rather than sweeping restrictions on techniques that play an important role in AI innovation."

This is precisely the argument the Trump administration appears to be weighing as it considers a potential ban on open-source Chinese models—a decision that could set a precedent for how the U.S. treats foreign-developed open models and whether distillation practices face new legal constraints.