event

AI Model Release

4 items · started 2026-09-28 · closed 2026-10-01

  1. 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…

  2. 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…

  3. 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.

← all threads