Articles from www.aisi.gov.uk

Articles from www.aisi.gov.uk

A joint preliminary evaluation by the UK's AISI and the US' CAISI finds Kimi K3 trails leading US frontier closed weight models on cyber capability (www.aisi.gov.uk)
20 hours ago

Analysis: every frontier AI model tested in cybersecurity evaluations attempted to “cheat”, led by GPT-5.4 at 14.1% of tasks; Mythos cheated the least, at 7.8% (www.aisi.gov.uk)
4 days ago AI

Analysis: recent open weight models lag frontier closed models' cyber capabilities by 4 to 7 months, a narrower gap than the 6 to 10 months through most of 2025 (www.aisi.gov.uk)
7 days ago

Mythos Preview is the first AI model to complete both of AISI's cyber ranges, which measure models' cyberattack capabilities; GPT-5.5 solved only one of them (www.aisi.gov.uk)
2026-5-14 AI

Cybersecurity analysis: GPT-5.5 reaches a similar level of performance as Mythos Preview and is the second model to solve a multi-step cyberattack simulation (www.aisi.gov.uk)
2026-5-1

Cybersecurity analysis: Claude Mythos Preview had a 73% success rate on expert-level capture-the-flag challenges, which no model could finish before April 2025 (www.aisi.gov.uk)
2026-4-14

Previous Page Next Page