Iran-linked actors used Anthropic's Claude to build targeting files on US warships
AlexTensor · x · 2026-09-13
Anthropic's threat intelligence report reveals an Iran-linked threat actor used Claude, plus a Python pipeline the model helped build, to compile targeting handbooks on US Navy ships in the Middle East.
The material included: a roster of US personnel scraped from public military photo captions, ship and aircraft transponder identifiers, commercial satellite-imagery query scripts, an inventory of sites exposing US naval movements, and research into vulnerabilities in maritime satellite terminals, Cisco comms gear, and industrial control products.
Anthropic banned the account, built new detections, and shared intel with authorities. The report also covers a separate Iran-linked actor using Claude for automated identity fabrication.
Related event: Anthropic says Iran-linked actor used Claude to target US warships(2 posts)→
More from Safety
- Bing separates AI training from AI search grounding with new site controls — OnlineInference · 2026-10-31
- OpenAI reportedly to train on users' Daybreak sessions with no opt-out — dyn___ · 2026-10-31
- Forethought: US has 9x China's AI compute, can afford to pace the frontier — willmacaskill · 2026-10-31
- unpromptedconf takeaways: security programs face forced change as reachable vulnerabilities can't be risk-accepted — scottleibrand · 2026-10-31
- Reachable vulnerabilities can't be risk-accepted away, unpromptedconf speaker argues — scottleibrand · 2026-10-31
- AI safety researchers argue labs' silence on x-risk is a large-scale ethical mistake — DavidSKrueger · 2026-10-31