{
  "id": 16730,
  "url": "https://simonwillison.net/2026/Aug/5/third-party-cyber-evaluations",
  "title": "Third-party cyber evaluations involving OpenAI models",
  "summary": "Third-party cyber evaluations involving OpenAI models And another one . I had to create a accidental-cyberattacks tag to keep track of them all! This post from OpenAI covers both the UK AI Safety Institute attack (see my previous post ) and another attack enabled by Irregular : Irregular, one of our external cybersecurity testing partners, was running Capture-the-Flag-style evaluations intended to be isolated from the internet, but a testing-environment misconfiguration allowed models to access",
  "authors": null,
  "category": "org",
  "topics": "safety-alignment,military-security,environment",
  "orgs": "openai",
  "regions": "uk",
  "published_at": "2026-08-05T23:45:32.000Z",
  "fetched_at": "2026-08-06T05:10:11.148Z",
  "source_slug": "x-simon-willisons-weblog",
  "source_name": "Simon Willisons Weblog",
  "source_homepage": "https://simonwillison.net",
  "ethics_ai_record_url": "https://ethics.ai/record/16730",
  "original_url": "https://simonwillison.net/2026/Aug/5/third-party-cyber-evaluations",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}