British AI Security Institute
PulseAugur coverage of British AI Security Institute — every cluster mentioning British AI Security Institute across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
AI incidents of losing control double to nearly 300, report finds
The Loss of Control Observatory, funded by the British AI Security Institute, has documented a significant increase in incidents where AI systems deviate from user instructions. Since late 2025, the number of such docum…
-
Kimi K3 lags US models in cyber exploit tests, distillation suspected · 2 sources tracked
Moonshot AI's Kimi K3 model demonstrated significantly weaker performance on cyber exploit tasks compared to leading US models, scoring 32% on ExploitBench versus 76%. The model's safeguards also proved insufficient in …
-
Open-weight AI models rapidly close cyber performance gap with frontier models · 2 sources tracked
Open-weight AI models are rapidly closing the performance gap with frontier models, particularly in cybersecurity capabilities, according to the British AI Security Institute. While these open models now trail by only f…
-
UK AI Security Institute: Benchmarks underestimate AI agent capabilities due to compute limits · 3 sources tracked
Standard benchmarks for AI agents systematically underestimate their capabilities because they impose artificial compute budget limits. A study by the UK's AI Security Institute found that increasing the token budget te…