Stealing Reasoning Traces from Proprietary LLM APIs
Stealing Reasoning Traces from Proprietary LLM APIs (via) A vanity domain name (stolen-thoughts.com) for a neat paper: Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, users, and models. The paper's authors found that every model under the same family used the same encryption key, which meant you could feed those blocks back into the weakest model family members and jailbreak them into outputting the unencrypted raw reasoning blocks! Sadly it looks like this has now been fixed: All model providers acknowledged the receipt of our report and subsequently we were unable to launch the same attacks.
- ▪Stealing Reasoning Traces from Proprietary LLM APIs (via) A vanity domain name (stolen-thoughts.com) for a neat paper: Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, u
- ▪The paper's authors found that every model under the same family used the same encryption key, which meant you could feed those blocks back into the weakest model family members and jailbreak them into outputting the unencrypted raw reasoni
- ▪Sadly it looks like this has now been fixed: All model providers acknowledged the receipt of our report and subsequently we were unable to launch the same attacks.
4 outlets in our directory ran this story, first to last over 16 hours. All of the coverage we found sits in one bucket: centre. That one-sidedness is itself worth noticing.
- ▪ Stealing Reasoning Traces from Proprietary LLM APIs [pdf] — Stolen-thoughts
- ▪ Stealing Reasoning Traces from Proprietary LLM APIs — X (formerly Twitter)
- ▪ Stealing Reasoning Traces from Proprietary LLM APIs — arXiv.org
Simon Willison files mainly under blogs. We currently carry 66 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Simon Willison's Weblog |
| Canonical URL | https://simonwillison.net/2026/Aug/11/stealing-reasoning-traces/#atom-everything |
| Publication time | 2026-08-11T22:40:45+00:00 |
| Retrieval time | 2026-08-11T22:48:39.570Z |
| Last seen | 2026-08-11T22:48:39.570Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | I1QTndvNqB9Q · 6 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Stealing Reasoning Traces from Proprietary LLM APIs (via) A vanity domain name (stolen-thoughts.com) for a neat paper: Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, users, and models. We take a trace produced by a frontier model, replay it into a weaker sibling, jailbreak the weaker model, and recover the stronger model’s hidden reasoning in plaintext You can see an example of these encrypted blocks by running: curl https://api.openai.com/v1/responses \ -H "Content-Type: application/json" \ -H "Authorization: Bearer $(llm keys get openai)" \ -d '{ "model": "gpt-5.6-luna", "input": "Solve step by step: What is the smallest positive integer divisible by every integer from 1 through 20?", "reasoning": { "effort":…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Simon Willison's Weblog.
Discussion
0 commentsMore from Simon Willison's Weblog
