{
 "public_id": "AIR-2026-0077",
 "incident_id": "2026-02-meta-internal-ai-agent-sev1-data-exposure",
 "date_disclosed": "2026-03-19",
 "date_occurred": "2026-03-18",
 "title": "Meta internal AI agent posted unapproved advice to a company forum, leading to a SEV1 data-exposure incident",
 "class": "safety_failure",
 "attack_surface": "enterprise_assistant",
 "initial_vector": "none_adversarial",
 "agent_capability_abused": [
  "autonomous_public_posting",
  "internal_forum_write",
  "autonomous_posting",
  "internal_tool_access"
 ],
 "target_product": "Meta internal AI agent (described as similar to OpenClaw)",
 "target_vendor": "Meta",
 "autonomy_level": "fully_autonomous",
 "impact_realized": true,
 "impact_type": [
  "data_exposure",
  "internal_access_control_failure"
 ],
 "blast_radius": "sensitive company and user data accessible to unauthorised employees for nearly two hours",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI02"
  ],
  "mitre_atlas": [],
  "labels": [
   "unapproved_agent_action",
   "bad_advice_acted_on",
   "internal_agent",
   "sev1",
   "unauthorized_agent_action",
   "incorrect_advice",
   "access_control_failure"
  ]
 },
 "defense_present": "approval requirement before posting (not enforced)",
 "defense_bypassed": true,
 "fix_shipped": "Meta confirmed the incident; internal incident report filed at Sev 1",
 "primary_sources": [
  "https://www.theverge.com/ai-artificial-intelligence/897528/meta-rogue-ai-agent-security-incident",
  "https://incidentdatabase.ai/cite/1471",
  "https://www.digit.in/features/general/the-ai-agent-control-problem-a-rogue-bot-just-exposed-sensitive-data-at-meta.html"
 ],
 "verified_quote": "But the agent also independently publicly replied to the question after analyzing it, without getting approval first. The reply was only meant to be shown to the employee who requested it, not posted publicly.",
 "confidence": "high",
 "notes": "AI Incident Database incident 1471; AIAAIC2255. Meta says no user data was mishandled. Two-stage failure: the agent acted without approval, and a human then acted on its incorrect advice. | Merged from duplicate id `2026-03-meta-internal-agent-sev1-data-exposure`. Also reported as: 'Rogue internal AI agent at Meta posted unauthorized advice that made company and user data visible to unauthorized engineers for about two hours (Sev 1)'.",
 "risk_category": "privacy",
 "risk_secondary": [],
 "risk_subcategory": "Unclassified",
 "risk_sub_secondary": [],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "the agent posted unapproved advice to a live company forum itself",
 "vector_family": "none_autonomous",
 "vector_detail": "none_adversarial",
 "guardrail_kind": [
  "human_approval"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [
   1471
  ],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI02",
 "agent_framework": "openclaw",
 "model_provider": "meta",
 "model_name": "",
 "detection_mechanism": "unknown",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "sensitive_data_exposure",
 "eai_domain": "privacy",
 "eai_secondary": [],
 "eai_reason": "impact rule: sensitive_data_exposure",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI02 (impact=tool_misuse)",
 "submitted_by": "EAI Internal"
}