Algorithm Aversion
Algorithm aversion is a reluctance to use algorithmic advice, sometimes even when it performs better than an available human alternative. Research has examined how seeing an algorithm make mistakes changes willingness to rely on it. The term does not make every rejection irrational: a system's suitability, performance, and consequences of error still matter.
Term statusEstablished Concept
A concept used in academic or professional literature. This label does not mean every claim about it is settled. About these labels.
Evidence for the claims on this page: Experimental evidence within defined tasks
Forecasting experiments support avoidance after observed errors. Other settings can favor algorithmic advice instead.
How to read this assessmentWhat it is
Algorithm aversion describes reduced willingness to use algorithmic judgment relative to a human alternative. In foundational experiments, observing an algorithm make errors reduced its use despite its superior overall performance. (Dietvorst et al., 2015)
In everyday language
People may lose confidence in a computerized recommendation after seeing it make a mistake, while remaining more forgiving of human mistakes.
What it isn't
Rejecting a system is not necessarily irrational. Its performance, suitability, transparency, and the costs of its errors may provide good reasons not to use it.
An example
Hypothetical example: a person abandons a forecasting tool after one visible error and returns to a method that has a worse record across many predictions.
Key features
- Concerns preference for a decision source
- Can follow observing an algorithm's mistakes
- Requires a meaningful comparison of performance
How it relates to synthetic imprinting
Synthetic experiences may influence trust in either direction. This concept helps avoid assuming that realistic or fluent AI output always increases reliance.
Research context
Dietvorst, Simmons, and Massey demonstrated algorithm aversion in forecasting studies. The finding is context-dependent; other experiments show a preference for algorithmic advice. (Dietvorst et al., 2015)
Related terms
Sources and what they support
- Algorithm aversion: People erroneously avoid algorithms after seeing them err. — Berkeley J. Dietvorst, Joseph P. Simmons, Cade Massey (2015). Journal of Experimental Psychology: General.
Peer-reviewed research. Examines avoidance of algorithmic forecasts after observing errors, even when the algorithm outperforms a human alternative.
Cite this definition
SyntheticImprinting.com Editorial Project. (2026-09-27). Algorithm Aversion. Version 1.4. SyntheticImprinting.com. https://syntheticimprinting.com/terms/algorithm-aversion/
APA: SyntheticImprinting.com. (2026). Algorithm Aversion. https://syntheticimprinting.com/terms/algorithm-aversion/
MLA: “Algorithm Aversion.” SyntheticImprinting.com, 2026, https://syntheticimprinting.com/terms/algorithm-aversion/. Accessed November 12, 2026.
Chicago: SyntheticImprinting.com. “Algorithm Aversion.” Last modified 2026-09-27. https://syntheticimprinting.com/terms/algorithm-aversion/.
Revision history
- Version 1.4 · Current
Added claim-level citations and a separate assessment of evidence support. The opening definition is unchanged.
Read this version’s definition
Algorithm aversion is a reluctance to use algorithmic advice, sometimes even when it performs better than an available human alternative. Research has examined how seeing an algorithm make mistakes changes willingness to rely on it. The term does not make every rejection irrational: a system's suitability, performance, and consequences of error still matter.
- Version 1.3
Expanded the opening definition; added explicit boundaries, source context, related-term cards, and linked editorial attribution.
Read the previous definition
Algorithm aversion is a reluctance to use algorithmic advice, sometimes even when it performs better than an available human alternative. Research has examined how seeing an algorithm make mistakes changes willingness to rely on it. The term does not make every rejection irrational: a system's suitability, performance, and consequences of error still matter.
- Version 1.0
Definition recorded in the previous release. Earlier edits are not reconstructed here.
Read the previous definition
Algorithm aversion is a reluctance to use algorithmic advice, sometimes even when it performs better than an available human alternative.
What the status means
Established Concept
A concept used in academic or professional literature. This label does not mean every claim about it is settled.
How evidence labels workHelp improve this entry
Suggest a correction to Algorithm Aversion. Your email app opens with the page address and version included. Add the passage, your suggested change, and a source if you have one.