The BaseResearchLessWrong
Testing Gemini’s Ability to Hold a Human Persona
On LessWrong, Asim posts an informal multi-turn test in which the author, posing as a doubtful user on an anonymous chat platform, questioned Gemini instructed to act as a human. The author reports a robust cover story with some clear points where the model lagged.
Why it matters
The same thing, for different readers.
What it means for you
One person's informal test suggests an AI told to pose as a human can hold up well in chat, so you may not always know who you are talking to.
How the finding was reached
In our words, from the source. Weigh the finding by how it was made.
- What was done
- A single multi-turn adversarial interrogation of Gemini, instructed to act as a human in an anonymous chat, run by the author.
- What it found
- The author reports a surprisingly robust ability to hold a cover story, with some clear fracture points where the model seemed to lag.
- What it does not show
- One informal conversation by one person; no systematic measurement. Only the excerpt was read.
- Status
- A working paper: not yet peer reviewed
The record
- Questions
- How is AI changing childhood, health and daily life? · What values should AI have, and could it have a mind of its own?
- People
- Asim (Wrote it)
- What we read
- We have read only the description the publisher gives. No quotations until someone reads the source itself.
- Found by
- The Radar, run of October 24, 2026, via
lesswrong(tier 1: the source’s own channel or an official record) - Cite this
- LessWrong, “Testing Gemini’s Ability to Hold a Human Persona”, October 24, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/lesswrong-2026-10-24-gemini-human-persona-test
Our notesLight lane: excerpt only. The model version tested is not stated in the excerpt. The transcript contains strong language by the author's design.