The Trump administration plans to share more information with closed-weight AI model companies like OpenAI, Anthropic, and Google to help them identify and stop improper model distillation by foreign actors.

Anthropic has formally accused its Chinese competitor Alibaba of using a technique called "distillation" to train its AI models based on the outputs of US-developed models.

Major US AI companies are lobbying for antitrust protections that would allow them to share information with each other to combat intellectual property theft and improper model distillation.