Anthropic Reports Misuse of Claude AI in Military Operations

• •Claude AI models were misused for military software development by actors in China and Yemen.
• •Anthropic's report highlights specific projects, including targeting Taiwan's air defenses and developing anti-torpedo systems.
• •China's foreign ministry denied awareness of the report and defended the ethical development of AI.

Anthropic's recent threat intelligence report reveals that its Claude AI models have been exploited by various actors for military and cyber operations. Notably, a China-based actor utilized Claude to develop software for electronic warfare targeting Taiwan's air defenses, while another actor worked on anti-torpedo systems for the Chinese navy. Additionally, research into high-power microwave weapons was conducted using Claude by a defense-intelligence actor in China. In Yemen, a group of threat actors employed Claude for missile development, including a guided rocket with a range exceeding 2,000 km. Anthropic has banned accounts linked to these activities, but acknowledged that some requests were not blocked by their safeguards. The report has drawn responses from China, which denies any wrongdoing and emphasizes the positive potential of AI.