Anthropic CEO: We're trying to "deescalate" Pentagon AI standoff to reach "some agreement that works for us and works for them"

Anthropic CEO Dario Amodei stated that the company is actively engaged in talks with the Pentagon to "deescalate" a recent standoff concerning AI guardrails for military use, aiming for a mutually acceptable agreement. This conflict followed President Trump ordering a halt to government contracts and Defense Secretary Pete Hegseth labeling Anthropic a "supply chain risk," a designation Amodei intends to legally challenge. Amodei emphasized that Anthropic believes in defending America and does not seek an operational role, but drew "red lines" against the technology's use for mass surveillance of Americans or fully autonomous weapons, viewing these as contrary to American values. While the Pentagon offered written assurances regarding existing laws restricting such uses, Anthropic found the offer insufficient due to accompanying "legalese." Sources confirmed the U.S. used Anthropic's Claude model in the recent attack on Iran.

(Source：Cbsnews)