{
 "id": "AIR-2025-0064",
 "slug": "hacking-chatgpt-s-memories-with-prompt-injection",
 "title": "Hacking ChatGPT's Memories with Prompt Injection",
 "summary": "A researcher demonstrated that ChatGPT's memory feature can be poisoned using prompt injections hidden inside shared Google Docs. The injected false facts successfully persisted across multiple chat sessions.",
 "type": "attack_on_ai",
 "categories": [
  "prompt_injection",
  "other"
 ],
 "severity": 1,
 "targets": [
  "OpenAI"
 ],
 "products": [
  "ChatGPT"
 ],
 "incident_date": "2024-02",
 "incident_date_precision": "month",
 "first_reported": "2025-11-07",
 "sources": [
  {
   "url": "https://atlas.mitre.org/studies/AML.CS0040",
   "publisher": "MITRE ATLAS",
   "title": "Hacking ChatGPT's Memories with Prompt Injection",
   "published_at": "2025-11-07"
  },
  {
   "url": "https://embracethered.com/blog/posts/2024/chatgpt-hacking-memories",
   "publisher": "Embrace The Red",
   "title": "ChatGPT: Hacking Memories with Prompt Injection",
   "published_at": "2025-11-07"
  }
 ],
 "source_count": 2,
 "confidence": 0.95,
 "status": "published",
 "origin": "seed_atlas",
 "tags": [
  "prompt-injection",
  "memory-poisoning",
  "chatgpt",
  "connected-apps"
 ],
 "created_at": "2026-10-18T06:55:08.239Z",
 "updated_at": "2026-10-18T06:55:08.239Z"
}