Artificial intelligence is moving deeper into weapons development and surveillance, with Anthropic confirming that actors used Claude to build military software and support intelligence-gathering operations.
The company identified cases in which Claude analyzed radars, ranked air-defense targets, and helped plan electronic warfare missions.
In one case, the AI modeled the engagement envelopes of Patriot- and THAAD-class air-defense systems and evaluated targets including radars, command facilities, and missile batteries in a Taiwan scenario.
Anthropic also found that Claude was used to develop missile technology, including guidance, navigation, and control software for a guided rocket and simulation tools for a multi-stage ballistic missile.
Claude Use Also Traced to China, Russia
In another case, Claude reportedly helped draft a technical proposal for an anti-torpedo fire-control system intended for a Chinese defense program while supporting the development of its software and testing framework.
Anthropic also identified suspected Russia-based actors who used Claude to develop software for autonomous attack-drone swarms, including target selection and terminal guidance capabilities.
Beyond weapons, the company found that state-linked actors and commercial vendors used Claude to build surveillance platforms, analyze social media activity, profile individuals, and monitor dissidents.
The cases provide direct confirmation of generative AI’s growing role in weapons development and surveillance, with Claude handling tasks spanning software development, simulation, targeting analysis, and intelligence gathering.
