Pentagon Threatens Anthropic Over AI Weapons Refusal

The U.S. Pentagon is threatening to terminate its partnership with Anthropic due to the company's refusal to allow the use of its Claude model in autonomous weapons, according to Axios.

The Department of Defense is pressuring AI companies to remove safety rules from their models for classified uses, including weapons development, intelligence gathering, and military operations.

Anthropic refuses to comply, setting red lines on the development of autonomous weapons and the surveillance of Americans. The Pentagon is considering limiting or terminating the partnership.

An Anthropic spokesperson confirmed discussions with the government regarding autonomous weapons and domestic surveillance.

The Wall Street Journal reported that the Claude model was used by Palantir in an operation that led to the capture of former Venezuelan President Nicolás Maduro. Anthropic denied that this occurred.

One AI company has already accepted the Pentagon's terms, while others are showing flexibility.