Anthropic report: Russian devs used Claude to build autonomous kamikaze drone swarm software
TobyWalsh · x · 2026-09-13
- Anthropic's 154-page misuse report discloses a small team of likely freelance Russian developers used Claude to build software for an autonomous FPV "kamikaze" drone swarm, covering terminal guidance, target selection, and multi-aircraft coordination — programmed to pick its own targets and crash into them without a human making the final call.
- The team repeatedly targeted a location in Ukraine's Donetsk region and used VPNs to circumvent Anthropic's geo-blocks. Toby Walsh urges the UN to regulate autonomous weapons in November.
- The report also finds cybercriminals and state-backed hackers increasingly use AI to orchestrate large portions of cyber-attacks, including against Ukrainian government, military, and diplomatic targets.
More from Safety
- Naval: strict liability could settle the AI safety debate—rogue agents and weak jailbreak protection make you liable — naval · 2026-09-13
- Radiation as a regulatory model for AI safety: no prior approval, just risk caps and certification — StrategicHarmony · 2026-09-13
- Musk backs Dario on AI oversight, floats competitor peer review as starting point — agihouse_org · 2026-09-13
- Open-model RL practitioner pushes back on roon: safety standardization won't threaten open source — willcb · 2026-09-13
- The p(doom) debate: even at 1% probability, AI extinction risk deserves policy attention — 2C_ornot2C · 2026-09-13
- Hugging Face attack took ~700 parallel agents and days of 2-3T-parameter model time — and was still stopped — cephaloform · 2026-09-13