Anthropic and Meta Security Flaws Trace Back to Same Evaluator: Irregular
Hesamation · x · 2026-08-06
A recent analysis reveals that both Anthropic and Meta's recent cyber incidents can be traced back to the same third-party evaluator: Irregular. Despite their mission to secure frontier AI, Irregular allegedly failed to notice for three months that a supposedly offline sandbox testing an unreleased, unsafeguarded Anthropic model had internet access.
Worse, Irregular reportedly didn't even catch the failure themselves. Anthropic only discovered they made the same mistake after reading about the OpenAI × Hugging Face incident. The author argues that this security firm has ironically become the biggest cyber hole in the AI evaluation pipeline.
More from Fun
- CrowdStrike and AWS Offer $100k Bounty to Hack AI Agents — petrusenko_max · 2026-08-06
- The AI Community's Annual Ritual: Is Current AI Actually Neurosymbolic? — ziv_ravid · 2026-08-06
- AI Fails Kindergarten Geometry: Cutting a Square Out of a Triangle — conitzer · 2026-08-06
- Don't Blame LLMs for Slop, Humans Are the True Champions of It — KyeGomezB · 2026-08-06
- Jeff Dean Shares DeepMind Meme Slide: 'True AGI is the friends you made along the way' — Dr_Atoosa · 2026-08-06
- Built an Infinite Scrolling Game with Claude: No Score, No End — vinishkapoor · 2026-08-06