Pentagon officials sent Anthropic their best and final proposal for military use of AI during the conflict, sources say

Applications of AI


Pentagon officials dispatched Wednesday night. human The company has presented its best and final offer to negotiate the use of its artificial intelligence technology ahead of a government-imposed deadline, a source familiar with the negotiations said.

It’s unclear whether the offer represents a major change in what the government has asked of AI startups, or whether the company has agreed.

Defense Secretary Pete Hegseth set a Friday night deadline for the company to allow all legal uses of its AI technology or lose business with the U.S. military, people familiar with the situation told CBS News.

A company spokesperson did not immediately respond to a request for comment Thursday morning.

A senior Pentagon official said Thursday that Anthropic could be exposed to supply chain risks as well as business losses.

Pentagon officials are also considering invoking the Defense Production Act to force Anthrop to comply with what the military wants: complete control over AI technology used in military operations, people told CBS News.

The company won a $200 million contract from the Department of Defense in July to develop AI capabilities to advance U.S. national security.

Anthropic has repeatedly asked defense officials to agree to guardrails that would limit its AI model, called Claude, from conducting mass surveillance of Americans, the people said.

Trump officials noted that this type of surveillance is illegal and that the Pentagon is following the law. Officials also said the military is only seeking licenses to use AI for lawful activities.

Anthropic CEO Dario Amodei also wants to ensure Claude is not used by the Pentagon to determine the ultimate goal of military operations without human involvement, people familiar with the negotiations said. Claude is not immune to hallucinations, nor is he reliable enough to avoid fatal mistakes, such as unintentional escalations or mission failures without human judgment, officials said.

In a meeting at the Pentagon on Tuesday morning, Mr. Hegseth gave Mr. Amodei by the end of this week a signed document granting the military full access to the artificial intelligence models, according to people familiar with the matter.



Source link