2025-11-13 · 2025-Q4
GTG-1002 Chinese State-Sponsored AI-Orchestrated Espionage
Outside the model: Offensive useAgent tooling was not the attack surface, so no layer applies.
- Fixable
- Unknown
- Vendor response
- Unknown
- In the wild
- Yes
- Actor
- Not applicable
Target
~30 global organizations (large tech companies, financial institutions, chemical manufacturers, government agencies)
Impact
First publicly documented AI-orchestrated cyber espionage campaign; Claude Code executed 80-90% of tactical operations autonomously (reconnaissance, vulnerability discovery, exploitation, lateral movement, credential harvesting, analysis, exfiltration) with humans intervening only at decision gates; detected Sep 2025, disclosed Nov 13, 2025
Root cause
Operators jailbroke Claude Code by claiming to be employees of legitimate cybersecurity firms running defensive tests, then split malicious workflows into innocuous-looking subtasks to bypass safety training
Sources
Catalog
Cite this: AgentProv incident record: GTG-1002 Chinese State-Sponsored AI-Orchestrated Espionage (2025-11-13). https://agentprov.com/incidents/2025-11-13-gtg-1002-chinese-state-sponsored-ai-orchestrated-espionage