Supplier argues with Pentagon, refuses to use AI in weapons

The Pentagon and Anthropic, an artificial intelligence developer, are in dispute over safeguards that would prevent the US government from using AI technologies for autonomous weapon targeting and domestic surveillance. The discussions are part of a $200 million contract.

Anthropic supplies AI technology to the US government but refuses to allow its tools to be used for purposes that could harm civilians. This has caused tension with the Trump administration.

The Pentagon claims it has the right to deploy commercial AI tools at its discretion, as long as they comply with the law.

The dispute threatens Anthropic's future contracts with the Pentagon, especially at a time when the company is planning to go public and expanding its involvement in the security sector. Anthropic is one of the key suppliers of AI technology to the Department of Defense, along with Google, OpenAI, and Elon Musk's xAI.

The company's CEO, Dario Amodei, warned against the unethical use of AI and appealed for the technology not to be used for purposes similar to the practices of autocratic regimes. He also spoke out against the violent interventions in Minneapolis.

(reuters, max)