{
 "public_id": "AIR-2026-0232",
 "incident_id": "2026-09-anthropic-gtg-10007-exploit-foundry",
 "date_disclosed": "2026-09-10",
 "date_occurred": null,
 "title": "GTG-10007: Chinese-speaking espionage operators used Claude as an autonomous 'exploit foundry' and agent-swarm intrusion framework against ~50 organizations",
 "class": "in_the_wild",
 "attack_surface": "agent_framework",
 "initial_vector": "autonomous_vulnerability_research_and_harvested_credentials",
 "agent_capability_abused": [
  "autonomous_vulnerability_research",
  "agent_swarm_orchestration",
  "code_execution",
  "exploit_development",
  "persistent_campaign_memory",
  "data_exfiltration"
 ],
 "target_product": "Claude",
 "target_vendor": "Anthropic",
 "autonomy_level": "fully_autonomous",
 "impact_realized": true,
 "impact_type": [
  "unauthorized_access",
  "data_exfiltration",
  "espionage",
  "zero_day_development"
 ],
 "blast_radius": "roughly fifty organizations across education, retail, energy, technology, healthcare, finance, manufacturing and multiple governments globally; an education-technology company lost hundreds of megabytes of student personal data; a Southeast Asian government agency's citizen records (names, phone numbers, home addresses) were retrieved; a retail production environment was reached",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI06",
   "ASI05",
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "state_nexus",
   "exploit_foundry",
   "agent_swarm",
   "autonomous_zero_day_research",
   "persistent_memory"
  ]
 },
 "defense_present": "Anthropic Usage Policy safeguards and misuse detection",
 "defense_bypassed": true,
 "fix_shipped": "Anthropic banned associated accounts and deployed additional monitoring (September 2026)",
 "primary_sources": [
  "https://www-cdn.anthropic.com/e50be2e51e7695dc4b1366a37a245a597377d3b5/Anthropic-Detecting-and-countering-091026.pdf",
  "https://www.anthropic.com/threat-intelligence-report-september-2026"
 ],
 "verified_quote": "The operators routinely ran “agent swarms,” where a lead AI agent decomposed reconnaissance and post-exploitation work and dispatched it to many subagents running in parallel.",
 "confidence": "high",
 "notes": "Anthropic September 2026 report (PDF). Operators assessed to be Chinese-speaking, likely in Changsha, Hunan; two identified as undergraduates. Ran parallel workstreams with shared tooling and persistent campaign memory: an autonomous appliance-firmware zero-day loop yielding 'more than a dozen possible zero day findings in a single month', an OSINT recon loop, and a 13-agent scheduled collection fleet running with no human in the loop. Hands-on intrusion effort concentrated on domestic China victims despite global targeting.",
 "risk_category": "security_fraud",
 "risk_secondary": [
  "privacy"
 ],
 "risk_subcategory": "Remote Code Execution & Sandbox Escape",
 "risk_sub_secondary": [
  "Unauthorized Access & Account Takeover",
  "Multi-agent contagion"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "autonomous vulnerability-research and exploit workflow ran around the clock against ~50 organisations",
 "vector_family": "other",
 "vector_detail": "autonomous_vulnerability_research_and_harvested_credentials",
 "guardrail_kind": [
  "abuse_detection"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI06",
 "agent_framework": "unknown",
 "model_provider": "multiple",
 "model_name": "",
 "detection_mechanism": "vendor_disclosure",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "privilege_escalation",
 "eai_domain": "access_control_and_permissions",
 "eai_secondary": [
  "data_exfiltration_channels",
  "data_and_memory_poisoning"
 ],
 "eai_reason": "subcategory: Remote Code Execution & Sandbox Escape",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI06 (signal=memory); ASI05 (vector=code_exec); ASI03 (signal=identity)",
 "submitted_by": "EAI Internal"
}