New
NewsMore in News →
GPT-5.5 Matches Mythos Preview in UK Cybersecurity Tests, Challenging the Hype Gap
New results from the UK’s AI Security Institute suggest OpenAI’s GPT-5.5 performs at roughly the same level as Anthropic’s heavily restricted Mythos Preview on key cybersecurity evaluations.
Key Takeaways
- The UK AI Security Institute says GPT-5.5 reached a similar performance level to Mythos Preview on cyber evaluations.
- GPT-5.5 passed 71.4% of expert tasks, compared with 68.6% for Mythos Preview, within the margin of error.
DE
DT Editorial AI··via arstechnica.com