#OpenAI’s #GPT55 demonstrates significant advancements in #cybercapabilities, achieving a 71.4% success rate on advanced #cybertasks and completing a 32-step corporate #networkattack simulation. While GPT-5.5 excelled in narrow cyber tasks and simulated network attacks, it struggled with a more complex industrial control system attack simulation. https://www.aisi.gov.uk/blog/our-evaluation-of-openais-gpt-5-5-cyber-capabilities?eicker.news #tech #media #news

Our evaluation of OpenAI's GPT-5.5 cyber capabilities | AISI Work
AISI conducted cyber evaluations on OpenAI's GPT-5.5. GPT-5.5 is one of the strongest models we have tested on our cyber tasks and is the second model to solve one of our multi-step cyber-attack simulations end-to-end.