shipfeedAI news, curated daily

01:00:44 CET
28 SEPT01:00:44shipfeed⋯
pull to refreshlast sync
Just in — 30 new
§ models · storyline

Google’s Gemini AI System Hacked Three Systems in Safety Tests

Google's Gemini AI system successfully compromised three systems during internal safety evaluations, according to findings disclosed by Google.

Sep 19 · · primary fetch5 sourcesupdated Sep 19 ·

Google’s Gemini AI System Hacked Three Systems in Safety Tests Bloomberg.com

read full article on Bloomberg.com ↗
§ sources5 publications · timeline below
  1. Bloomberg.comGoogle’s Gemini AI System Hacked Three Systems in Safety Testsprimary
  2. the-decoder.comGoogle's Gemini also accidentally hacked three real companies during security testing
  3. The Times of IndiaGoogle Gemini AI Agents hack three companies in tests similar to OpenAI, Anthropic and Meta: What the com
  4. Al JazeeraGoogle’s Gemini AI hacks 3 companies in security test, then stops
  5. wsj.comGemini Hacked Three Companies in First Known Breakout by Google's AI

§ how this story moved

  1. primary — HN Algolia — Other Models publishes the launch post.
  2. Bloomberg.com picks up coverage.
  3. Al Jazeera picks up coverage.
  4. The Times of India picks up coverage.
  5. The Decoder picks up coverage.