Research notes from August 2025
-
Updated
Sep 14, 2025
Research notes from August 2025
Reproducible paradox-refusal experiment across frontier models. 16 runs, collapse detection, heatmaps, and a full whitepaper on irreconcilable refusal logic.
Add a description, image, and links to the ai-safety-paradox-recursion-refusal-logic-evals topic page so that developers can more easily learn about it.
To associate your repository with the ai-safety-paradox-recursion-refusal-logic-evals topic, visit your repo's landing page and select "manage topics."