Conforming to AI vs. humans: The moderating effects of social exclusion and task type
HeeJun Lee, Hyeontaek Oh, Yoonkyung Jeong · 2026
Read Conforming to AI vs. humans: The moderating effects of social exclusion and task type by HeeJun Lee, Hyeontaek Oh, Yoonkyung Jeong (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-0355478.txt.
Description
HeeJun Lee, Hyeontaek Oh, and Yoonkyung Jeong run an online experiment on whether people conform to opinions from AI as they do to opinions from humans. Several hundred participants judged a series of tasks. Some tasks had clear right answers. Some were matters of taste. Each judgment followed a stated opinion sourced either from an AI or from people. The team also measured how socially excluded each participant felt.
Feeling excluded reduced conformity to the AI's opinions but left conformity to human opinions untouched. Task type, objective versus subjective, did not change the source effect. Lee and colleagues read the split as a sign about motives. Following AI rests more on judging its usefulness as information. Following people rests more on social pull. The gap widens when people feel disconnected.
Catalog record
| CREATOR | HeeJun Lee, Hyeontaek Oh, Yoonkyung Jeong |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32206 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |