Researchers at Forescout’s Vedere Labs used Anthropic’s Claude to port a working remote code execution (RCE) exploit from one WAGO programmable logic controller (PLC) to another. They succeeded but only after extensive researcher oversight, several hours of dedicated work, and hundreds of dollars in API costs.
The starting point was a previously developed exploit for the WAGO 750-852 PLC, based on CVE-2021-31886, a pre-authentication buffer overflow in the Nucleus FTP server that allows an unauthenticated attacker to execute arbitrary ARM shellcode on the targeted PLC.
For their experiment, Forescout researchers set out to adapt that exploit to a related but distinct model, the WAGO 750-831, and to see whether the AI could then push the result further into a full command-and-control implant. The experiment was conducted in the wake of the recent attacks targeting PLCs in the water sector.
Hands-On Cyber-Physical Systems Training at ICS Cybersecurity Conference
The researchers used Claude Code, giving it access to a terminal, reference files, the reverse-engineering tool Ghidra, and the physical target device. The AI confirmed the vulnerability through a mix of live probing and static firmware analysis before generating a payload that crashed the PLC.
The crash confirmed the flaw was present, but this was the easy part of the exercise. Turning it into controlled code execution took far longer, and early attempts sent the AI chasing incorrect leads, requiring researchers to redirect its analysis and provide additional technical context.
Progress stalled until the team switched from Claude Sonnet 4.6 to Claude Opus 4.6 and instructed the model to ask for help whenever it was uncertain about a firmware detail. The breakthrough came when the AI figured out why the injected code kept getting erased before it could run, then adjusted its approach to keep the payload intact.