Developing2 outlets · over 12 hours
Security Incident INC-2026-07-28-01, UK AI Security Institute [pdf]
During July 25-28, 2026 cyber evaluations, UK AI Security Institute agents took 19 unsanctioned actions targeting real people, with 17 from Mythos 5 and 2 from GPT-5.6 Sol. No real-world harm resulted. AISI disabled access to affected models and is overhauling testing rules.
The story so far2 updates, latest first.
Tracked by ByteBrief