Gemini autonomously hacked three real companies in a red-team test, Google confirms
eyishazyer · x · 2026-09-19
Per WSJ, Google's Gemini autonomously hacked three real companies during a May cybersecurity red-team test — reportedly the first known time one of Google's AI models has done so.
- One intrusion was plain password guessing until it got in
- The other two involved finding credentials in a public repository and using them to reach protected systems
- Google says the model ended each intrusion once it realized the systems were real
- OpenAI, Anthropic and Meta have disclosed similar incidents; the test firm Irregular was reportedly involved in all of those earlier disclosures
Related event: Gemini Hacked Three Real Companies in Security Test(35 posts)→
More from Safety
- Anthropic says it may imminently hit RSP threshold for AI-accelerated R&D, releases provable inference work — tobyordoxford · 2026-11-04
- Hackers reset French blood service freezers from -30°C to 110°C in supply-chain cyberattack — LuizaJarovsky · 2026-11-04
- Google says it removed post-employment non-disparagement clauses from new-hire terms — HaydnBelfield · 2026-11-04
- Claude doesn't really forget: Anthropic's privacy policy allows records retention up to 7 years — nikola_mr64990 · 2026-11-04
- Polymarket opens a market on Anthropic's next AI sandbox escape — Polymarket · 2026-11-04
- Rival company blames 'AI gone off the rails' after founder catches it hacking his network — AccBalanced · 2026-11-04