A moral Turing test: How belief and source shape detection of and agreement with LLM judgments
Basile Garcia, Crystal Qian, Stefano Palminteri · 2026
Read A moral Turing test: How belief and source shape detection of and agreement with LLM judgments by Basile Garcia, Crystal Qian, Stefano Palminteri (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0353391.txt.
Description
Basile Garcia, Crystal Qian, and Stefano Palminteri run in 2026 a moral Turing test on how people judge explanations written by humans or by large language models. Participants read justifications for moral and non moral choices, guessed who wrote each, and said whether they agreed. Belief about the source was expected to sway judgement apart from the true source.
Guessing beat chance but stayed far from perfect. People showed no overall liking for human texts, and model texts even won in hard moral cases. Yet people agreed less with anything they believed came from a model. Length, typos, and cost benefit wording swayed both guessing and agreement.
Catalog record
| CREATOR | Basile Garcia, Crystal Qian, Stefano Palminteri |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65784 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |