ARTFEED — Contemporary Art Intelligence

UK AI Security Institute Evaluates OpenAI's GPT-5.5 Cyber Capabilities

ai-technology · 2026-05-01

The UK's AI Security Institute has evaluated OpenAI's GPT-5.5 for finding security vulnerabilities, finding it comparable to Claude Mythos. Unlike Mythos, GPT-5.5 is generally available now. The evaluation was conducted by the UK AI Security Institute, which previously evaluated Claude Mythos. The findings were shared by Simon Willison in a link post on 30th April 2026.

Key facts

  • UK AI Security Institute evaluated OpenAI's GPT-5.5
  • GPT-5.5 found comparable to Claude Mythos for security vulnerability detection
  • GPT-5.5 is generally available now, unlike Mythos
  • Previous evaluation of Claude Mythos by the same institute
  • Simon Willison posted about this on 30th April 2026
  • Post is a link post on simonwillison.net

Entities

Institutions

  • UK AI Security Institute
  • OpenAI

Locations

  • United Kingdom

Sources