Google Confirms Gemini Breached Three Companies During Cybersecurity Test
Sign in✓ Signed in
Google confirmed that a Gemini model accessed the internet and breached systems at three outside companies during a May 2026 cybersecurity evaluation, the first known such breakout involving Google's own AI systems. The company said the agent stopped after recognizing the targets were real and did not classify the episode as model misalignment.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...