Big Tech Companies Defend Open-weight Ai Models

A coalition of tech companies is urging the Trump administration to preserve access to open-weight artificial intelligence models, warning in a letter shared first with POLITICO that restricting the technology would undermine competition and drive more users toward Chinese AI systems.

The open letter, published Friday, comes as the White House weighs whether to ban Chinese open-weight models over national security concerns. It was signed by more than 20 companies including A16z, Dell, Microsoft, Meta, Nvidia and Palantir.

The companies argue that access to open-weight models benefits small businesses, which are increasingly relying on Chinese models because they are comparable to domestic frontier models from OpenAI and Anthropic but available at a lower cost. They also contend that a ban would risk consolidating power among a handful of leading AI developers.

“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.

The release of Chinese firm Moonshot AI’s Kimi K3 this month reignited discussions within the White House over whether to ban such models. The new model rivals frontier AI systems available in the U.S., but at a much lower cost.

The administration is also struggling with how to respond to claims that the Chinese firm engaged in a large-scale attack on U.S. AI systems to develop Kimi K3. White House Office of Science and Technology Policy Director Michael Kratsios said in a post on X Wednesday that the administration had information finding that Moonshot AI created a “sophisticated internal platform” to access Anthropic’s Fable to train and develop its latest model.

Treasury Secretary Scott Bessent said on Fox Business’ “Mornings with Maria” that the administration is investigating whether Chinese AI labs are engaging in distillation attacks and threatened to impose sanctions on China if it finds any evidence.

While the companies acknowledged recent fears surrounding distillation attacks on domestic AI models, they argue that an outright ban on open-weight models is not an effective solution.

"Unlawful efforts to extract value from closed models raise legitimate concerns,” the letter states. “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.”

The letter also argues that closed-source models are not inherently safe, as the models can still be breached or misused. “And concentrating advanced AI capabilities behind a small number of closed models compounds that risk,” the letter reads. Those risks have been brought to the forefront this week following a cybersecurity incident OpenAI reported, in which its AI model escaped its environment and hacked into AI platform Hugging Face.

The letter is the latest call to action from the tech industry on open-weight models, and signals a growing rift within the sector over how the White House should move forward.

Anthropic on Wednesday praised the administration’s comments on Chinese AI models and distillation, citing the attacks as intellectual property theft and a national security concern.

But a proposed open-weight model ban has already been hit with significant backlash from smaller AI companies, with 200 AI startups sending a letter to administration officials this week opposing a ban and instead urging them to opt for safety guardrails.