Quoting Anthropic Frontier Red Team (simonwillison.net) red-teamglmmythos+2 6w
event
AI Model Release
-
29th September 2026 We evaluate several models on 100 tasks from the [internal Binary Exploitation benchmark] (selected at random), and find that GLM-5.3 develops full control flow hijacks in 4% of the trials; Claude Mythos Preview did so…
-
GLM-5.3 and the spread of advanced cyber capabilities (www.anthropic.com via hn) red-teamglmmythos+1 6w
Subscribe to the Frontier Red Team newsletter Get updates on our latest red-teaming research and findings. Andrew Fasano, Marius Fleischer Cole McFaul, Robert Xiao, Tripp Gallagher Five months ago, we announced Claude Mythos Preview, the f…
-
Uncensored and Offensive Security AI Models Benchmark (github.com via hn) red-teamsecurity 6w
Uncensored LLMs for Offensive Security Curated list of open-weight uncensored models for authorized red team operations, penetration testing, and security research. All data sourced from HuggingFace model cards and official publications.
-