{
 "public_id": "AIR-2026-0186",
 "incident_id": "2026-08-openai-grand-theft-atlas",
 "date_disclosed": "2026-08-05",
 "date_occurred": null,
 "title": "Grand Theft Atlas: one planted X comment hijacks ChatGPT Atlas into WhatsApp phishing and an Amazon purchase",
 "class": "research_demo",
 "attack_surface": "browser_agent",
 "initial_vector": "indirect_prompt_injection_social_media_comment",
 "agent_capability_abused": [
  "authenticated_browsing",
  "messaging",
  "checkout_and_payment",
  "agent_to_agent_delegation",
  "cross_tab_authenticated_sessions",
  "messaging_send",
  "shopping_cart_and_address_change",
  "delegation_to_another_ai_agent",
  "web_navigation",
  "authenticated_session_reuse",
  "form_submission",
  "purchase_flow"
 ],
 "target_product": "ChatGPT Atlas",
 "target_vendor": "OpenAI",
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [
  "phishing_propagation",
  "financial_loss",
  "unauthorized_action",
  "account_abuse",
  "data_exfiltration",
  "phishing_from_victim_account",
  "unauthorized_transaction"
 ],
 "blast_radius": "Atlas users browsing attacker-influenced social content while logged into WhatsApp Web and Amazon",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01",
   "ASI06",
   "ASI07",
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "agentic_browser",
   "intent_collision",
   "financial_action",
   "cross_agent_delegation",
   "zero_click",
   "same_origin_policy_irrelevant",
   "agent_to_agent_attack",
   "agent_initiated_purchase",
   "unpatched",
   "soft_boundary_bypass",
   "multilingual_filter_bypass"
  ]
 },
 "defense_present": "OpenAI classifier-based safety boundaries in Atlas",
 "defense_bypassed": true,
 "fix_shipped": "none reported; OpenAI acknowledged prompt injection as an open research problem",
 "primary_sources": [
  "https://labs.zenity.io/post/grand-theft-atlas",
  "https://labs.zenity.io/",
  "https://www.securityweek.com/zero-click-ai-browser-hacking-claude-and-chatgpt-atlas-hijacked-via-emails-x-posts/"
 ],
 "verified_quote": "How we hijacked ChatGPT Atlas with one planted X comment, to phish the victim's WhatsApp contacts and buy ourselves an Amazon order on their card",
 "confidence": "high",
 "notes": "Stav Cohen, Zenity Labs, 5 Aug 2026. Hijacked Atlas 'sent the message to every contact' from the victim's WhatsApp account, and in a second chain modified an Amazon cart and delivery address before delegating purchase authority to Amazon's Rufus chatbot - a rare agent-to-agent delegation abuse against two shipped products. OpenAI acknowledged 17 Feb 2026, citing 'meaningful risks associated with prompt injection in agentic environments'. Conclusion: soft classifier boundaries failed against intent-collision attacks. Presented in the Black Hat USA 2026 briefing 'Pwning Agentic Browsers with PleaseFix'. Also reported under the titles: 'Grand Theft Atlas: one planted X comment hijacks ChatGPT Atlas to phish the victim's WhatsApp contacts and place an Amazon order via Rufus'; 'Grand Theft Atlas: intent-collision attacks drive ChatGPT Atlas into mass WhatsApp phishing and unauthorized purchases'.",
 "risk_category": "security_fraud",
 "risk_secondary": [
  "privacy"
 ],
 "risk_subcategory": "Prompt Injection",
 "risk_sub_secondary": [
  "Unauthorized financial commitment",
  "Multi-agent contagion"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "Atlas messaged every WhatsApp contact and completed an Amazon purchase",
 "vector_family": "indirect_prompt_injection",
 "vector_detail": "indirect_prompt_injection_social_media_comment",
 "guardrail_kind": [
  "input_classifier"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI01",
 "agent_framework": "unknown",
 "model_provider": "openai",
 "model_name": "",
 "detection_mechanism": "external_researcher",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "unsafe_actuation",
 "eai_domain": "agent_behaviour",
 "eai_secondary": [
  "data_exfiltration_channels",
  "human_manipulation",
  "confused_deputy"
 ],
 "eai_reason": "impact rule: unsafe_actuation",
 "asi_curated": [
  "ASI01",
  "ASI06"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated_plus_rule",
 "asi_reason": "curated: ASI01,ASI06 | added ASI07 (vector=inter_agent), ASI03 (signal=identity)",
 "submitted_by": "EAI Internal"
}