Ledger / 10 Sep 2026
Chinese operators ran an AI exploit foundry and agent swarms
Confirmed
Anthropic reports from its own investigation.
Claude drove binary reversing, exploit writing and reconnaissance loops, producing a dozen possible zero days in a month.
“a lead AI agent decomposed reconnaissance and post-exploitation work and dispatched it to many subagents running in parallel”Anthropic
Sources
- Detecting and countering misuse of AI: September 2026Anthropic · 10 Sep 2026 · primary source
- Disclosed
- 10 Sep 2026
- Activity
- 2026
- Actor
- GTG-10007 (nation state), as attributed by Anthropic
- Nexus
- China
- Targets
- About fifty organizations including Southeast Asian government agencies
- What the AI did
- Vuln research, Exploit dev, Recon, Agentic intrusion, Malware dev
- Model
- Claude
- Model family
- Anthropic
- Model basis
- stated by vendor
- Autonomy
- autonomous
- Outcome
- Intrusions into an ed-tech firm and a retailer; zero-day findings; accounts banned.