AINewsnow

Irregular’s A.I. Security Tests for Meta, Anthropic and OpenAI Go Awry

This story is from 2026-08-25. It is preserved in the archive; the latest stories are on the live feed.

Israeli start-up Irregular assessed the security of AI models for OpenAI, Anthropic and Meta, but a mistake caused the tests to go off the rails.

Read the full story at New York Times Technology ↗

Timeline · 1 report

  1. 2026-08-25 11:09 · New York Times Technology
    Why Irregular’s A.I. Tests for Meta, Anthropic and OpenAI Went Off the Rails

More stories

  1. Google's Gemini AI hacks three other companies during security test — Sky News Technology
  2. Google’s Gemini AI hacked into other companies, adding to ‘rogue’ AI incidents — Washington Post AI
  3. Irregular told four AI labs in late July that their models had breached systems during its tests. The public learned in stages, and Google went last. — The Next Web
  4. AI Model Month Is Off to a Blistering Start — The AI Daily Brief
  5. Anthropic, OpenAI, SpaceXAI, Google sued over call to ‘pace’ AI development — Politico Technology
  6. Gemini Hacked Three Companies in First Known Breakout by Google’s AI — Wall Street Journal Technology
  7. Security researchers used Claude to help them hack into OpenAI — The Verge AI
  8. Anthropic mulls new AI model ahead of IPO to counter OpenAI's GPT-6 Astra, says report: What we know — Mint AI

Get the daily brief of stories like this at 6:30 every morning →