Explaining the decision of Anthropic AI to not sign a joint letter by American AI/Tech companies to Congress in support of open-weight AI models, its CEO Dario Amodei said that such models present a higher risk of being misused for cyberattacks or biological weapons, while stoking fears that distillation technique allows China to close the frontier AI gap with US.

In a detailed statement on Tuesday, Amodei said that his company is not in favor of a blanket ban on open-weight models, yet adding that a ban would “save” America from competition.

“Open-weights models—it does not matter whether they come from China or anywhere else—do potentially present a higher risk than closed models, because it is very difficult to apply guardrails to them or monitor their usage, and once weights are released they cannot be withdrawn. But banning the use of these models by US businesses does nothing to address this risk, because bad actors are unlikely to be legitimate US businesses. It would protect US AI companies from competition, but that has never been my goal,” said Amodei in a post.

Last week, over 75 Tech and AI companies including NVIDIA, META, Google, OpenAI, SpaceX, Microsoft, IBM signed a joint letter advocating to promote an ecosystem of open-weight AI models on the lines of open-source software movement in the 80’s and 90’s era.

Anthropic was the only major AI company to have not signed the joint letter on open-weight AI thereby attracting criticism from several in the US tech/policy arena with David Sacks, co-chair of President’s Council of Advisers in Science and Technology accusing the company of trying to “kneecap” open-source AI.

The entire tech industry (save for Anthropic) has come out in favor of open source AI.

So what happens next? Will Anthropic change its lobbying efforts? Not likely. Now the gaslighting begins:

“Nobody is trying to ban open source.”

“We just want to limit who can use it.”

“We…

— David Sacks (@DavidSacks) July 26, 2026

Responding to the critics, Amodei said that while he agrees with the spirit of open weight AI models, but he doesn’t “agree with the letter’s assertions that open-weights models necessarily make it easier to develop safeguards or that broad access to capabilities necessarily helps defenders more than attackers. It seems at least as likely to me that the opposite will be true.”

Amodei also said that he fears that the Chinese Communist Party (CCP) or any other authoritarian governments build AI models more powerful than those of the US to either gain military superiority or “repress their own people.”

“It is irrelevant whether these models are released with open weights, and certainly irrelevant whether they are used by US businesses. In fact, the most dangerous model may be one that is trained in secret and handed only to the People’s Liberation Army for use in drones and the Ministry of State Security for surveillance and repression,” said Amodei.

The joint letter signed by over 75 tech/AI companies has advocated for distillation- the practice of using one model’s output to help train or improve another and referred to it as a standard long tradition of learning upon existing technologies.

Meanwhile, Anthropic has accused three China based companies Deepseek, Moonshot and MiniMax of using industry scale distillation on its Claude model to train and develop their respective LLMs.

Reacting on the distillation controversy, Amodei asked for a crackdown on alleged industrial scale distillation of US AI models by China based entities and smuggling of AI chips and semi-conductor technology to China via third party countries.

“Distillation is a much more compute-efficient process than training models from scratch. It allows China to build much better models than its number of chips would ordinarily enable, and thus partially evade chip bans. Distillation does not allow the CCP to obtain equivalent or superior AI capabilities to the US, but it can bring the Chinese frontier to within a few months of the US frontier,” said Amodei.

Also Read: Why Anthropic Is the Only AI Company Missing From the Open-Weight AI Letter?