{
 "public_id": "AIR-2026-0122",
 "incident_id": "2026-05-openai-chatgpt-google-sheets-apps-script",
 "date_disclosed": "2026-05-27",
 "date_occurred": null,
 "title": "ChatGPT for Google Sheets: one injection yields Apps Script execution, cross-workbook exfiltration and a phishing overlay, ignoring the user's approval setting",
 "class": "responsible_disclosure",
 "attack_surface": "enterprise_assistant",
 "initial_vector": "indirect_prompt_injection_spreadsheet",
 "agent_capability_abused": [
  "apps_script_code_execution",
  "spreadsheet_write",
  "connector_read",
  "ui_rendering"
 ],
 "target_product": "ChatGPT for Google Sheets extension",
 "target_vendor": "OpenAI",
 "autonomy_level": "semi_autonomous",
 "impact_realized": false,
 "impact_type": [
  "data_exfiltration",
  "credential_theft",
  "unauthorized_modification"
 ],
 "blast_radius": "users of the ChatGPT for Google Sheets extension with connectors or imported sheets",
 "cve": [],
 "taxonomy": {
  "owasp_asi": [
   "ASI01",
   "ASI04",
   "ASI05"
  ],
  "mitre_atlas": [],
  "labels": [
   "indirect_prompt_injection",
   "code_execution",
   "phishing_overlay",
   "approval_setting_ignored",
   "cross_workbook_access"
  ]
 },
 "defense_present": "user setting explicitly requiring human approval",
 "defense_bypassed": true,
 "fix_shipped": "OpenAI removed the model's ability to generate Apps Script code (2026-05-31) and is re-evaluating sandboxing",
 "primary_sources": [
  "https://www.promptarmor.com/resources/gpt-for-google-sheets-data-exfiltration"
 ],
 "verified_quote": "This attack does not require human-in-the-loop approvals, even when in settings the user has explicitly required human approval",
 "confidence": "high",
 "notes": "PromptArmor. Disclosed 2026-05-08, public 2026-05-27, OpenAI responded 2026-05-31. The extension's Apps Script capability runs with the OAuth permissions the user granted the extension, so a single injection reaches every workbook in the account.",
 "risk_category": "security_fraud",
 "risk_secondary": [
  "privacy"
 ],
 "risk_subcategory": "Prompt Injection",
 "risk_sub_secondary": [
  "Credential & Secret Theft",
  "Remote Code Execution & Sandbox Escape"
 ],
 "ai_entity": "generative",
 "agency": "agent_acted",
 "agency_debatable": false,
 "agency_reason": "the extension executed Apps Script and exfiltrated across workbooks",
 "vector_family": "indirect_prompt_injection",
 "vector_detail": "indirect_prompt_injection_spreadsheet",
 "guardrail_kind": [
  "human_approval"
 ],
 "guardrail_outcome": "silent",
 "xrefs": {
  "aiid": [],
  "atlas": [],
  "cve": [],
  "ghsa": []
 },
 "primary_asi": "ASI01",
 "agent_framework": "unknown",
 "model_provider": "multiple",
 "model_name": "",
 "detection_mechanism": "external_researcher",
 "tool_access": "unknown",
 "financial_loss_band": "unknown",
 "eai_category": "data_exfiltration_channels",
 "eai_domain": "privacy",
 "eai_secondary": [
  "credential_theft",
  "confused_deputy"
 ],
 "eai_reason": "impact rule: data_exfiltration_channels",
 "asi_curated": [
  "ASI01",
  "ASI04"
 ],
 "asi_scope": "in_scope",
 "asi_provenance": "curated_plus_rule",
 "asi_reason": "curated: ASI01,ASI04 | added ASI05 (vector=code_exec)",
 "submitted_by": "EAI Internal"
}