Anthropic says it has shut down a cell of threat actors based in northern Yemen that used Claude to build guidance software for three weapons programmes running in parallel. The group was working on a guided rocket, a multi-stage ballistic missile with a stated range goal above 2,000km, and a multi-variant missile set called the R2000, which included a hypersonic glide vehicle. In its September threat report on detecting and countering misuse of AI, the company says the actors used Claude Code in place of human software engineers.The setup is the part that stands out. The cell did not simply ask an AI model for advice on a hard problem. It ran several Claude instances at the same time and assigned each one a role, the way a lead delegates work on a small engineering team. One instance wrote the code. A second handled research. A third reviewed what the first one produced.
Claude was used to write the software that steers and stabilises a flying vehicle
The AI model’s job was guidance, navigation and control. In one instance the actors used Claude to integrate an open-source autopilot onto a phone-class flight computer, the sort of commodity hardware found in consumer devices. Claude wrote the control and position estimation software, tuned the control settings, ran the firmware build pipeline and performed a flight simulation. The report also documents reinforcement learning used to tune flight control, and six degrees of freedom trajectory simulation on the ballistic missile side.Anthropic’s safeguards blocked many of the requests. Not all of them. The actors hid their goals and concealed what the software was meant for, then split tasks across many sessions so no single one revealed the full programme.
A failed rocket test sent Yemen terrorist group back to Claude within hours
Anthropic says it has no evidence the cell managed to field an operational device. It did test-fire a guided rocket, and that field test appears to have failed. The company knows this because the actors came back to Claude within hours to work out why, bringing post-test telemetry in for diagnosis.Northern Yemen is controlled by the Houthi rebels, who have been attacking shipping in the Red Sea and pushing toward the coast near the Bab el-Mandeb Strait. Anthropic does not name the users. A member of the Houthi political bureau, Hazam al-Assad, called it unreasonable and illogical to suggest the group would rely on open sources for military production, saying its forces have their own accumulated capabilities.
Banning the Yemen cell’s Claude accounts did not take the toolkit away
Anthropic banned every account it could link to the cell and shared threat information with public and private sector partners. The complication is what the group had already finished. Before the ban, it compiled a simulation toolkit into a standalone executable, a digital model of the weapons system that runs without Claude and without engineering environments such as MATLAB. That part of the work can continue with no platform access at all.The Yemen case is one of six conventional weapons operations detailed in the report, which covers activity Anthropic disrupted between December 2025 and August 2026. The company notes that this kind of programme has historically been pieced together by governments and United Nations panels from recovered hardware, long after the fact. Here, the software was visible while it was still being written.