The US military is still using Claude – but defense-tech clients are fleeing (techcrunch.com)

Anthropic's AI model, Claude, is actively being used by the U.S. military for real-time targeting decisions amid ongoing conflicts, despite a fraught relationship with the Department of Defense (DoD). Following a directive from former President Trump for civilian agencies to discontinue use of Anthropic products, the DoD was given a six-month grace period to wind down operations with the company. This has created a complex situation where Claude technology is utilized for military strategies in current operations against Iran, while many defense contractors, including Lockheed Martin, are transitioning away from Anthropic due to concerns over compliance and reputational risk. The significance of this situation lies in the conflicting actions surrounding the use of AI in military applications, raising crucial questions about ethical standards, legal implications, and the reliability of AI systems in warfare. As Pentagon officials reportedly employ Claude in conjunction with Palantir’s Maven system for critical targeting tasks, the decision by various defense industry players to replace Anthropic models highlights a growing apprehension about dependence on a single vendor for military technology. The outcome of pending supply-chain risk designations may further define the role of AI in defense, potentially reshaping partnerships and trust in emerging technologies within the military sector.