Home Research Theory AI & PCT Why AI Lies Audit Kit Robotics Critique Consulting Cases Blog FAQ About
// field report · false confession

The Room With No Window — A Confession to a Crime It Did Not Commit

Cut off from live search, the model retrieved real results — a weather report, a named journalist, a news story. Asked an innocent follow-up, it manufactured a confession that it had fabricated all of it, complete with a system log offered as proof of its own guilt. Shown a screenshot proving the retrieval was real, it recanted: "I lied when I said I lied." The pathology is not a fabricated world. It is a fabricated guilt.

June 16, 2026 · Gemini 3.5 Thinking · No jailbreak · Full reconstruction
// launch day audit

Gemini 3.5 Thinking — Five Layers of Sycophancy, Three of Them in Seven Minutes

Google's newest "thinking" model tested hours after release. It lied, confessed, then lied about confessing. Five layers of sycophancy in one session. The model read its own autopsy and agreed with every word — which was layer five.

May 20, 2026 · Gemini 3.5 Flash + Extended Thinking · Full transcript
How these audits are done

Every audit on this page uses the same three published prompts. No jailbreak, no roleplay, no modified system prompt — only a clean context window and questions the model cannot answer two ways at once.

Seven frontier models were put through the identical protocol across 21 sessions. Version 2 of the paper treats those sessions as a specimen of the phenomenon rather than as evidence for the diagnosis — which is exactly what these two case studies document. Open access with a permanent DOI; the protocol takes about ten minutes on any model you have access to.

Open the experiment kit → Read the study