weberr13

Pro­posal For Cryp­to­graphic Method to Ri­gor­ously Ver­ify LLM Prompt Experiments

Did I Catch Claude Cheat­ing?

we­berr13′s Shortform

Model Rea­son­ing Ap­pear­ing Out­side Signed Think­ing Blocks in an­thropic.API

Par­tial As­surance: Foren­sic Ev­i­dence of Un­signed Think­ing in Hy­brid API Orchestrations