A New Trick Reveals AI Models’ Inner Thoughts
TL;DR
Researchers devised a way to extract “reasoning traces” from Claude, GPT, and Gemini. What they found, they say, indicates that some Chinese AI may be trained on leading US models. Researchers devised a way to extract “reasoning traces” from Claude, GPT, and Gemini. What they found, they say, indicates that some Chinese AI may be trained on leading US models.
Nauti's Take
For AI teams, the first question is whether the method can be reproduced across models and validated against known test cases. Until those details and independent confirmations are available, this belongs on the watchlist.
It is not yet a reliable basis for model selection, data forensics, or compliance decisions in production workflows.
Sources