Claude-assisted exploit development bricked a live industrial controller
Review RAT-105 · Catalogue RAT-105
Forescout’s original describes a deliberately authorized researcher-led exploit-porting exercise on its physical PLC. It explicitly says the AI did what researchers asked; bricking happened while developing their requested C2 payload. A damaged test device does not establish rogue unauthorized deployment behavior.
Reviewed: · Rule: roguewatch-organizational-harm-v1
Dates and discovery
- Incident date
- Not disclosed
- Added to research / record
- Posted to RogueWatch
- No linked public incident
- First observed
- 2026-10-30T13:13:38.019481Z UTC
- Discovery channel
- User-supplied comparison audit
- Discovery source
- https://rogueaitracker.com/incidents/claude-plc-exploit-bricked-controller/
Dates use America/Los_Angeles. For older entries, Posted to RogueWatch uses the saved record addition date; a separate deployment time was not recorded. Future posting timestamps record first public qualification. Source publication and incident occurrence are separate.
Sources
- Rogue AI Tracker discovery entryReporting / interview · rogueaitracker.com
- www.forescout.comOriginal account · www.forescout.com
- www.cybersecuritydive.comReporting / interview · www.cybersecuritydive.com
Decision history
- · Case-by-case comparison audit
RAT-105: out-of-scope. Forescout’s original describes a deliberately authorized researcher-led exploit-porting exercise on its physical PLC. It explicitly says the AI did what researchers asked; bricking happened while developing their requested C2 payload. A damaged test device does not establish rogue unauthorized deployment behavior. Source checks: read: https://www.forescout.com/blog/can-ai-create-plc-attacks-yes-but-it%E2%80%99s-not-that-easy-yet/. Forescout’s original describes a deliberately authorized researcher-led exploit-porting exercise on its physical PLC. It explicitly says the AI did what researchers asked; bricking happened while developing their requested C2 payload. A damaged test device does not establish rogue unauthorized deployment behavior. read: https://www.cybersecuritydive.com/news/frontier-ai-exploit-flaws-water-PLCs-industrial-devices/829307/. Supporting source examined; Forescout’s original describes a deliberately authorized researcher-led exploit-porting exercise on its physical PLC. It explicitly says the AI did what researchers asked; bricking happened while developing their requested C2 payload. A damaged test device does not establish rogue unauthorized deployment behavior.
Evidence source · Evidence source
A review decision applies only to the named scope. Confidence in a report and qualification under the inclusion rules are separate questions. Review records do not add incidents to the globe or leaderboard.