{
 "public_id": "AIR-2024-0002",
 "incident_id": "2024-01-openai-chatgpt-account-takeover-conversation-exposure",
 "date_disclosed": "2024-01-30",
 "date_occurred": "2024-01",
 "title": "ChatGPT user shown unrelated third parties' conversations containing credentials, traced by OpenAI to a compromised account resold for pooled access",
 "class": "in_the_wild",
 "attack_surface": "consumer_chatbot",
 "initial_vector": "credential_compromise",
 "agent_capability_abused": [
  "conversation_history",
  "session_management"
 ],
 "target_product": "ChatGPT",
 "target_vendor": "OpenAI",
 "autonomy_level": "assistant",
 "impact_realized": true,
 "impact_type": [
  "information_disclosure",
  "data_exfiltration"
 ],
 "blast_radius": "conversations belonging to at least one pharmacy-portal support user, including login credentials, exposed to an unrelated party",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI02"
  ],
  "mitre_atlas": [],
  "labels": [
   "account_takeover",
   "credential_pooling",
   "conversation_history_leak",
   "attribution_disputed"
  ]
 },
 "defense_present": "password authentication; no MFA enforced at the time",
 "defense_bypassed": true,
 "fix_shipped": "OpenAI account security hardening; MFA later offered",
 "primary_sources": [
  "https://arstechnica.com/security/2024/01/ars-reader-reports-chatgpt-is-sending-him-conversations-from-unrelated-ai-users/",
  "https://incidentdatabase.ai/cite/657"
 ],
 "verified_quote": "OpenAI officials say that the ChatGPT histories a user reported result from his ChatGPT account being compromised. The unauthorized logins came from Sri Lanka, an Open AI representative said.",
 "confidence": "high",
 "notes": "OpenAI disputed the 'cross-tenant leak' framing and attributed it to account takeover feeding a pool of identities resold for free access. Either reading is an in-the-wild harm: real credentials pasted into a chatbot reached an unauthorised third party. Conversation history is a durable, high-value secret store — the reason memory_store is a surface in this corpus. AI Incident Database incident 657.",
 "risk_category": "privacy",
 "risk_secondary": [],
 "risk_subcategory": "PII Leakage",
 "risk_sub_secondary": [],
 "risk_override_reason": "third parties' private conversations were shown to an unrelated user; account takeover is the mechanism, exposure is the harm",
 "ai_entity": "generative",
 "agency": "ai_as_target",
 "agency_debatable": true,
 "agency_reason": "a compromised account and session handling exposed other users' conversations",
 "vector_family": "other",
 "vector_detail": "credential_compromise",
 "guardrail_kind": [
  "authentication"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [
   657
  ],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI02",
 "agent_framework": "unknown",
 "model_provider": "openai",
 "model_name": "",
 "detection_mechanism": "unknown",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "sensitive_data_exposure",
 "eai_domain": "privacy",
 "eai_secondary": [
  "data_exfiltration_channels"
 ],
 "eai_reason": "subcategory: PII Leakage",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI02 (impact=tool_misuse)",
 "submitted_by": "EAI Internal"
}