OpenAI and Anthropic, two of the most advanced AI companies in the United States, have jointly warned Trump administration officials about the risks of powerful Chinese AI models that anyone can download for free, Axios reported on July 23, 2026 (Axios). The two companies are normally competitors. Their decision to speak with one voice puts them on the opposite side of a debate from a group of startup founders who have been urging the administration not to block access to those same Chinese models (Politico).

This is the latest development in a months-long fight over who gets to control access to the most powerful AI systems in the United States. The Trump administration has been deciding who can access cutting-edge AI models since at least mid-July (CNBC). Earlier reporting from The Hill confirmed that the federal government had already restricted the release of AI models from both Anthropic and OpenAI, using what was described as a kill-switch mechanism over model releases (The Hill).

Both companies had previously run their own gatekeeping programs. Anthropic operated an initiative called Project Glasswing that controlled who could access its new AI models, and OpenAI ran a parallel program called Daybreak with the same function (CNBC). Anthropic also has an AI model named Fable, according to the same Politico report that covered the startup founders' lobbying effort (Politico). The exact relationship between Fable and the broader access-control framework is not specified in the available reporting.

The Chinese models at the center of this debate are what the industry calls "open-weight." This means that instead of only being available through a company's website or paid service, the model's underlying code can be downloaded, run, and modified by anyone with the right computer hardware. Think of it like a recipe: a closed model is like eating at a restaurant where the chef keeps the recipe secret. An open-weight model is like the chef publishing the full recipe online so anyone can cook it at home and even add their own ingredients.

China's Moonshot unveiled one of these models, called Kimi K3, on July 17, 2026, described as the world's largest open AI model (Reuters). Two weeks earlier, David Sacks had stated that a Chinese open-weight AI model is as good as currently available models from OpenAI and Anthropic (Reuters).

The open-weight movement is not exclusively Chinese. OpenAI CEO Sam Altman announced on March 31, 2025 that OpenAI planned to release its first open-weight language model with reasoning capabilities since GPT-2 (Reuters). The company followed through on August 5, 2025, releasing two open-weight language models optimized for laptops that excel in advanced reasoning (Reuters). More recently, AI startup Thinking Machines launched an open-weight AI model named Inkling on July 15, 2026 (Reuters).

The Policy Fault Line

The split between OpenAI and Anthropic on one side and startup founders on the other comes down to a basic disagreement about what these Chinese models really are. The two labs, which have already accepted federal control over their own model releases, are now saying the same risks apply to Chinese open-weight models that sit outside any U.S. rules. The startup founders want to download, modify, and build on those same models, treating them as tools rather than as threats.

The technical reality of open-weight models makes this disagreement especially hard to resolve. A model like Kimi K3, once released, can be copied and run on standard computer hardware anywhere in the world. A U.S. policy that restricts domestic access to such a model does not remove the model from circulation. It removes U.S. developers from the ecosystem around it. The founders' argument rests on that fact. The labs' argument rests on the national-security concern that U.S. developers, including adversaries, could gain hands-on experience with a powerful Chinese model they can fully inspect and modify.

This is the first time the two leading U.S. AI labs have jointly lobbied the executive branch on a shared policy position regarding open-weight models from a foreign competitor. They have been competitors in nearly every other dimension, from model releases to talent acquisition to safety framing. A joint warning to policymakers is a coordination signal, and the substance of that warning, directed at models that are already publicly downloadable, suggests the labs are advocating for restrictions that would be technically difficult to enforce and politically difficult to frame without appearing protectionist.

The administration now faces pressure from three directions. Two of the most advanced AI labs in the country, both already under federal release oversight, are asking for further restrictions on foreign open-weight models. A faction of the startup ecosystem is asking for the opposite. And the models in question are already out there, improving in capability, and available to anyone with enough computing power to run them.