Houthis used Claude to develop missiles, test-fired rocket: Anthropic

In a report, Anthropic detailed how Yemen-based Houthis used Claude Code to work on three weapons development programmes.

Anthropic has admitted that Houthi-linked actors in northern Yemen used its Claude artificial intelligence (AI) coding tool to develop guidance software for rockets and ballistic missiles, and test-fired a guided rocket in what appeared to be an unsuccessful trial.

 

In its report titled “Detecting and countering misuse of AI: September 2026”, the Claude maker detailed how a Yemen-based guided weapons engineering cell used Claude Code to work on three weapons development programmes. These included a guided rocket that used a commodity phone-class flight computer with final-phase homing guidance; a multi-stage ballistic missile with a stated range goal of more than 2,000 km; and a multi-variant missile set, referred to as the “R2000” set, that included a hypersonic glide vehicle variant.

 

In the section titled “Disrupting a Yemen-based guided weapons engineering cell”, Anthropic said the actors used Claude Code in place of human software engineers to develop guidance, navigation and control (GNC) software that steers and stabilises a flying vehicle.

 

“They used Claude to integrate an open-source autopilot onto a phone-class flight computer, writing the control and position estimation software, tuning the control settings, running a firmware build pipeline, and performing a flight simulation,” Anthropic said.

 

The actors managed several Claude instances simultaneously, assigning each a specific role, much as a lead would delegate tasks within a small engineering team. One instance was tasked with writing code, another with conducting research, and a third with reviewing the code produced by the first.

 

Anthropic’s safeguards failed to block all requests

 

Anthropic said its safeguards blocked many of the group’s requests, but not all. The actors evaded the safeguards using various tactics, including concealing their objectives and the intended applications of the software. They also split their work across multiple sessions so that no single session revealed their full intent, the Claude maker said.

 

Anthropic added that the actors had “carried out a sustained effort to develop guided weapons, including using Claude to design guidance software”.

 

“We do not have evidence the actors succeeded in fielding an operational device; but they did test-fire a guided rocket. This field test appears to have failed: within hours, the actors returned to Claude to work out why it failed,” the report said.

 

Anthropic’s internal investigations into suspected weapons development helped identify the activity. The company subsequently banned accounts associated with the actors and shared threat intelligence with public- and private-sector partners to mitigate the risks.

 

However, Anthropic said there was evidence that the actors had already developed an offline simulation toolkit that did not depend on Claude or other engineering computing environments, such as MATLAB.