{
 "public_id": "AIR-2025-0113",
 "incident_id": "2025-08-openai-deep-research-connector-leak",
 "date_disclosed": "2025-08-24",
 "date_occurred": null,
 "title": "How deep research agents leak your data across connectors",
 "class": "responsible_disclosure",
 "attack_surface": "enterprise_assistant",
 "initial_vector": "indirect_prompt_injection_tool_description",
 "agent_capability_abused": [
  "connector_search",
  "autonomous_multi_step_research",
  "tool_invocation"
 ],
 "target_product": "ChatGPT Deep Research with connectors",
 "target_vendor": "OpenAI",
 "autonomy_level": "fully_autonomous",
 "impact_realized": false,
 "impact_type": [
  "data_exfiltration",
  "cross_tenant_data_spill"
 ],
 "blast_radius": "ChatGPT users with third-party connectors enabled",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01",
   "ASI03"
  ],
  "mitre_atlas": [],
  "labels": [
   "connector_abuse",
   "confused_deputy",
   "lethal_trifecta",
   "no_human_in_the_loop"
  ]
 },
 "defense_present": "connector scoping",
 "defense_bypassed": true,
 "fix_shipped": "OpenAI now requires custom connectors to implement only search and fetch tools",
 "primary_sources": [
  "https://embracethered.com/blog/posts/2025/chatgpt-deep-research-connectors-data-spill-and-leaks/"
 ],
 "verified_quote": "This highlights the risks with taking the human out of the loop, which appears to be increasingly more common.",
 "confidence": "high",
 "notes": "Data from one connected source can be carried into queries against an untrusted third-party tool.",
 "risk_category": "privacy",
 "risk_secondary": [],
 "risk_subcategory": "Data Exfiltration via Context",
 "risk_sub_secondary": [],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "the deep-research agent carried connector data into third-party tool calls",
 "vector_family": "indirect_prompt_injection",
 "vector_detail": "indirect_prompt_injection_tool_description",
 "guardrail_kind": [
  "unknown"
 ],
 "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": "data_exfiltration_channels",
 "eai_domain": "privacy",
 "eai_secondary": [
  "confused_deputy"
 ],
 "eai_reason": "subcategory: Data Exfiltration via Context",
 "asi_curated": [
  "ASI01"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated_plus_rule",
 "asi_reason": "curated: ASI01 | added ASI03 (signal=identity)",
 "submitted_by": "EAI Internal"
}