Particle.news
Download on the App Store

Pentagon Reviews Anthropic Ties After Reports Claude Was Used in Caracas Operation

A Pentagon review puts Anthropic’s $200 million engagement at risk over the company’s usage limits.

Overview

  • Axios reports officials are considering labeling Anthropic a supply chain risk, a move that could force other defense contractors to sever relationships with the AI firm.
  • The Pentagon confirmed its relationship with Anthropic is under review, with a spokesperson saying partners must enable troops to prevail in any fight.
  • The Wall Street Journal reported Claude was used through a Palantir integration in a JSOC mission in Caracas, though neither Anthropic nor the Department of Defense has confirmed operational use.
  • Anthropic said it cannot comment on specific operations and reiterated that any deployment must follow its Usage Policies, which restrict facilitation of violence, weapons design, and surveillance.
  • Claude was the first frontier model on classified Pentagon networks under an engagement worth up to $200 million, as the department presses AI vendors including Google, OpenAI, and xAI to allow all lawful uses.