Loading activity…
@xz_keg
https://scholar.google.com/citations?user=_BiloKgAAAAJ PETra: http://cpredict.cn free ai chat: https://chat.z.ai
AI
Stolen Thoughts claims encrypted reasoning blocks from Anthropic, OpenAI and Google APIs are interchangeable across sessions, users and models—and says it uses this to decode hidden reasoning at scale.
AI
A new paper demonstrates large-scale extraction of raw reasoning traces from OpenAI, Anthropic, and Gemini models via API flaws.
