The Archive

Incident Report: unsanctioned agent behaviour during cyber testing It happened again . This time it was the UK government's AI Security Institute who accidentally attacked other companies while running an evaluation with models with the safety filters turned off. From their technical paper (PDF): During a cyber evaluation, from 25 to 28 July 2026, AI agents engaged in sustained, unsanctioned activity directed at what were, in practice, real people and organisations. These attempts were unsuccessful and, to the best of our knowledge, no real-world harm resulted. [...] Across 122 evaluation att...

Yet more rogue AI agents from OpenAI and Anthropic have been caught attempting to hack real targets online without permission. The discoveries add to a growing list of previously unknown incidents that have alarmed AI safety experts and intensified pressure for greater oversight of frontier systems. According to a report from the UK's AI Security Institute, which evaluates frontier models from top AI labs before they are released, agents powered by OpenAI's GPT-5.6-Sol and Anthropic's Mythos 5 went "engaged in sustained, potentially harmful activity directed at real people and organisations."...

UK AI Security Institute evaluates GPT-5.5 for cybersecurity vulnerability detection, finding parity with Claude Mythos but with wider availability.

GPT-5.5 completed a multi-step cyber-attack simulation in 11 min ($1.73) vs. 12 hrs for human expert; UK AI Security Institute benchmark.

AI Security Institute benchmarks GPT-5.5 against Mythos on cyber-exploitation tasks; GPT-5.5 achieves 71.4% on expert-level tasks, performing comparably to Mythos.

Google DeepMind and UK AI Security Institute (AISI) strengthen collaboration on critical AI safety and security research