AI-first company

Study

Colleagues judge people who use AI as less competent

People expected to be judged as less competent and less motivated if they used AI at work, and evaluators did judge AI users that way. The penalty also affected how evaluators rated job candidates.

By
Last reviewed
Type of study
Randomised experiment
Published
2025
Corrections
Report an error

At a glance

Source
Reif, Larrick and Soll (2025). Evidence of a social evaluation penalty for using AI. Proceedings of the National Academy of Sciences, 122(19), e2426766122. doi:10.1073/pnas.2426766122
How it was done
Four online experiments, registered in advance. Participants were assigned by chance to conditions in which a worker did or did not use AI, and either predicted how they would be judged or judged others. Randomised experiment. People were assigned by chance to work with or without the tool, and the two groups were compared.
Who took part
4,439 participants across four experiments.
Where
Online experiments run by researchers at Duke University's Fuqua School of Business, United States.

“Although AI can enhance productivity, its use carries social costs.”

What it found

  • People expect the penalty. Workers expected colleagues to rate them as less competent and less motivated if they used AI.
  • The penalty is real. Evaluators did rate people who used AI as less competent and less motivated.
  • It reaches hiring. The same judgements affected how evaluators assessed job candidates.

When the result holds

A worker who expects to be judged for using AI has a reason to use it quietly. A reviewer who does not know AI was used cannot check for the mistakes AI tends to make.

Limits

  • The studies are online experiments, not observation of real workplaces.
  • They do not measure how often people hide their AI use.

Practices that use this study

This study supports: