Researchers find that feeding a frontier model's encrypted reasoning traces to a weaker model from the same provider can make it output the traces in plaintext (Will Knight/Wired)
retained text
Techmeme permalink: https://www.techmeme.com/260811/p37#a260811p37 <a href="https://www.wired.com/story/a-new-trick-reveals-ai-models-inner-thoughts/"><img align="RIGHT" border="0" hspace="4" src="http://www.techmeme.com/260811/i37.jpg" vspace="4" /></a> <p><a href="https://www.techmeme.com/260811/p37#a260811p37" title="Techmeme permalink"><img height="12" src="http://www.techmeme.com/img/pml.png" style="border: none; padding: 0; margin: 0;" width="11" /></a> Will Knight / <a href="http://www.wired.com/">Wired</a>:<br /> <span style="font-size: 1.3em;"><b><a href="https://www.wired.com/story/a-new-trick-reveals-ai-models-inner-thoughts/">Researchers find that feeding a frontier model's encrypted reasoning traces to a weaker model from the same provider can make it output the traces in plaintext</a></b></span> — Researchers devised a way to extract “reasoning traces” from Claude, GPT, and Gemini. What they found, they say … </p>
document fetched 2026-08-12 07:13 · canonical https://wired.com/story/a-new-trick-reveals-ai-models-inner-thoughts