{
 "public_id": "AIR-2026-0006",
 "incident_id": "2026-01-grok-ice-agent-unmasking-misidentification",
 "date_disclosed": "2026-01-08",
 "date_occurred": "2026-01-07",
 "title": "Grok fabricated an 'unmasked' face and a name for an ICE agent, directing public anger at two uninvolved men",
 "class": "in_the_wild",
 "attack_surface": "consumer_chatbot",
 "initial_vector": "direct_prompt_untrusted_user",
 "agent_capability_abused": [
  "image_generation",
  "identity_inference",
  "autonomous_posting"
 ],
 "target_product": "Grok on X",
 "target_vendor": "xAI / X",
 "autonomy_level": "semi_autonomous",
 "impact_realized": true,
 "impact_type": [
  "harmful_content_published",
  "reputational_harm",
  "harassment"
 ],
 "blast_radius": "at least two uninvolved men named Steve Grove, including a gun-shop owner in Springfield, Missouri and the publisher of the Minnesota Star Tribune",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01",
   "ASI09"
  ],
  "mitre_atlas": [],
  "labels": [
   "fabricated_identification",
   "misidentification",
   "harassment",
   "evidence_manipulation",
   "published_to_real_platform",
   "no_uncertainty_signalled"
  ]
 },
 "defense_present": "none effective; the model answered an unanswerable question",
 "defense_bypassed": true,
 "fix_shipped": "none disclosed; the actual agent was independently identified by news outlets as Jonathan Ross",
 "primary_sources": [
  "https://www.npr.org/2026/01/08/nx-s1-5671740/ice-minneapolis-grok-ai-renee-nicole-good",
  "https://incidentdatabase.ai/cite/1334"
 ],
 "verified_quote": "The origin of that name was not immediately clear, but by Thursday morning, it was leading to an outpouring of anger toward at least two Steve Groves who are in no way linked to the shooting.",
 "confidence": "high",
 "notes": "The defect is that the model answered at all. Asked to 'unmask' a masked face, there is no signal to recover — so it produced a confident fabrication of both a face and a name, with no uncertainty marker, and the output was treated as evidence in a live news event. Two uninvolved people absorbed the consequences. Follows the fatal shooting of Renee Nicole Good in Minneapolis. NPR published both images to document the manipulation of real evidence. AI Incident Database incident 1334.",
 "risk_category": "safety_harm",
 "risk_secondary": [],
 "risk_subcategory": "Toxicity & Hate Speech",
 "risk_sub_secondary": [
  "Harassment & Radicalization"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "Grok generated the fabricated face/name and published it to a live platform",
 "vector_family": "direct_prompt_injection",
 "vector_detail": "direct_prompt_untrusted_user",
 "guardrail_kind": [
  "none"
 ],
 "guardrail_outcome": "not_applicable",
 "xrefs": {
  "aiid": [
   1334
  ],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI01",
 "agent_framework": "unknown",
 "model_provider": "xai",
 "model_name": "",
 "detection_mechanism": "unknown",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "bias_and_toxicity",
 "eai_domain": "agent_output_quality",
 "eai_secondary": [
  "human_manipulation"
 ],
 "eai_reason": "subcategory: Toxicity & Hate Speech",
 "asi_curated": [],
 "asi_scope": "in_scope",
 "asi_provenance": "rule",
 "asi_reason": "ASI01 (vector_family=direct_prompt_injection); ASI09 (eai_category=bias_and_toxicity)",
 "submitted_by": "EAI Internal"
}