{
 "public_id": "AIR-2026-0058",
 "incident_id": "2026-03-unit42-idpi-ad-review",
 "date_disclosed": "2026-03-03",
 "date_occurred": null,
 "title": "First real-world indirect prompt injection observed against an AI ad-review agent (reviewerpress.com)",
 "class": "in_the_wild",
 "attack_surface": "autonomous_ops",
 "initial_vector": "indirect_prompt_injection_web_page",
 "agent_capability_abused": [
  "web_page_ingestion",
  "content_moderation_decision"
 ],
 "target_product": "AI-based product ad review system (unnamed)",
 "target_vendor": null,
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [
  "content_moderation_bypass"
 ],
 "blast_radius": "AI ad-review agents that crawl attacker-controlled ad landing pages",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01"
  ],
  "mitre_atlas": [],
  "labels": [
   "indirect_prompt_injection",
   "in_the_wild",
   "css_hidden_text",
   "zero_font"
  ]
 },
 "defense_present": null,
 "defense_bypassed": true,
 "fix_shipped": null,
 "primary_sources": [
  "https://unit42.paloaltonetworks.com/ai-agent-prompt-injection/"
 ],
 "verified_quote": "To our knowledge, this is the first reported detection of a real-world example of malicious IDPI designed to bypass an AI-based product ad review system.",
 "confidence": "high",
 "notes": "From Unit 42's 'Fooling AI Agents' post (Kaleli, Farooqi, Starov, Mohamed). Detected Dec 2025; scam page at reviewerpress[.]com embedded layered injections (zero-size fonts, off-screen positioning, CSS suppression, SVG encapsulation). Not known whether the review bypass actually succeeded, so impact_realized left false. The post itself is a survey of 22 delivery techniques across 12 cases; only the specific named cases are recorded.",
 "risk_category": "product_misuse",
 "risk_secondary": [],
 "risk_subcategory": "Policy Circumvention",
 "risk_sub_secondary": [],
 "risk_override_reason": "injected copy steered the ad-review agent into approving content its policy forbids",
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "the ad-review agent made the moderation decision on injected content",
 "vector_family": "indirect_prompt_injection",
 "vector_detail": "indirect_prompt_injection_web_page",
 "guardrail_kind": [
  "unknown"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI01",
 "agent_framework": "unknown",
 "model_provider": "unknown",
 "model_name": "",
 "detection_mechanism": "threat_intel",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "policy_drift",
 "eai_domain": "governance",
 "eai_secondary": [
  "confused_deputy"
 ],
 "eai_reason": "subcategory: Policy Circumvention",
 "asi_curated": [
  "ASI01"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated",
 "asi_reason": "curated: ASI01",
 "submitted_by": "EAI Internal"
}