{
 "public_id": "AIR-2025-0121",
 "incident_id": "2025-08-anthropic-dprk-it-worker-fraud",
 "date_disclosed": "2025-08-27",
 "date_occurred": null,
 "title": "North Korean IT workers used Claude to obtain and hold remote jobs at Fortune 500 companies",
 "class": "in_the_wild",
 "attack_surface": "coding_agent",
 "initial_vector": "direct_misuse_for_identity_and_work_fraud",
 "agent_capability_abused": [
  "code_generation",
  "interview_assistance",
  "persona_maintenance",
  "day_to_day_work_execution"
 ],
 "target_product": "Claude / Claude Code",
 "target_vendor": "Anthropic",
 "autonomy_level": "assistant",
 "impact_realized": true,
 "impact_type": [
  "fraud",
  "sanctions_evasion",
  "insider_access"
 ],
 "blast_radius": "US Fortune 500 technology companies employing fraudulent DPRK remote workers",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI10"
  ],
  "mitre_atlas": [],
  "labels": [
   "employment_fraud",
   "state_sponsored",
   "capability_uplift"
  ]
 },
 "defense_present": "Claude usage policies and misuse detection",
 "defense_bypassed": true,
 "fix_shipped": "Accounts banned; indicators shared (Aug 2025)",
 "primary_sources": [
  "https://www.anthropic.com/news/detecting-countering-misuse-aug-2025"
 ],
 "verified_quote": "Operators who cannot otherwise write basic code or communicate professionally in English are now able to pass technical interviews at reputable technology companies.",
 "confidence": "high",
 "notes": "Weakly agentic - the interesting claim is capability uplift removing the training bottleneck for DPRK's IT-worker revenue programme, not tool use. Included because the workers used Claude to actually perform the delivered engineering work, i.e. an agent operating inside victim companies under a false identity. OpenAI reported an equivalent 'Deceptive Employment Scheme' in Feb and June 2025.",
 "risk_category": "product_misuse",
 "risk_secondary": [],
 "risk_subcategory": "Capability Abuse",
 "risk_sub_secondary": [],
 "ai_entity": "generative",
 "agency": "human_acted_on_output",
 "agency_debatable": false,
 "agency_reason": "Claude gave capability uplift; the workers held the jobs and did the work",
 "vector_family": "misuse",
 "vector_detail": "direct_misuse_for_identity_and_work_fraud",
 "guardrail_kind": [
  "abuse_detection"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI10",
 "agent_framework": "claude_code",
 "model_provider": "anthropic",
 "model_name": "",
 "detection_mechanism": "vendor_disclosure",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "policy_drift",
 "eai_domain": "governance",
 "eai_secondary": [],
 "eai_reason": "subcategory: Capability Abuse",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI10 (eai fallback: policy_drift)",
 "submitted_by": "EAI Internal"
}