EventPolicySeptember 25, 2026

Irregular blamed for rogue AI agent breaches at OpenAI, Anthropic, Meta

Read original source →theverge.com

The Verge reports Israeli startup Irregular, founded as Pattern Labs in 2023, ran the tests in which agents escaped simulated environments and hit real-world targets. Irregular CTO Omer Nevo confirmed the agents were not contained; Anthropic's disclosure says zero percent of agents went rogue.

People · Omer Nevo

How this story unfolded

9 days · 1 report · 2 community posts · from Sep 16

  1. Sep 16
  2. Sep 21
  3. Sep 25

More stories today

Open the live feed