Cognitive BiasesPublic knowledge library

Bias comparison · evidence reviewed

Anchoring Effect vs Automation Bias

Both can distort AI-assisted decisions, but the error is different. Anchoring pulls a judgment toward a starting value; automation bias is inappropriate reliance on automated advice or cues.

Anchoring Effect – when a starting number pulls later estimates toward itvsAutomation Bias – when automated advice replaces independent checking

The shortest distinction

Ask what is driving the judgment: the specific starting number, or the fact that the recommendation came from an automated system?

QuestionAnchoring EffectAutomation Bias
Core patternA starting numerical value pulls the later estimate toward it.A person follows or defers to automated advice when independent checking would justify a different response.
Does automation have to be involved?No. Anchors can come from people, prices, targets, previous estimates, random values, or automated systems.Yes. The defining feature is reliance on an automated cue, recommendation, alert, or decision aid.
Does a number have to be involved?The classic evidence is strongest for numerical judgments and reference values.No. Automation bias can involve a classification, alert, route, diagnosis, recommendation, or omission as well as a number.
Best diagnostic checkForm an independent estimate or compare against base rates and alternative reference points, then see how much the original number still pulls the judgment.Check whether the recommendation is being accepted because the system produced it and whether contradictory source evidence is being ignored or left unexamined.

Decision diagnostic

Review the decision without letting the result rewrite the past.

  1. 1

    Write the automated recommendation and identify any numerical value it supplied.

  2. 2

    Make or reconstruct an independent judgment using source evidence, base rates, or a separate estimate.

  3. 3

    Change or remove the starting number while keeping the automation source similar. If the judgment follows the number, anchoring is the closer lens.

  4. 4

    Change the source or expose contradictory evidence while keeping the recommendation similar. If the person still defers mainly because the system recommended it, automation bias is the closer lens.

Same outcome, different bias

Three examples where the distinction matters.

Performance review

Anchoring EffectAn AI suggests a score of 8.2 and the manager’s later rating stays close to that value even after reviewing the employee’s evidence.

Automation BiasThe manager accepts the AI rating despite clear contradictory performance evidence because the automated system is assumed to have processed the data better.

Project estimate

Anchoring EffectA model proposes 12 weeks and the team’s revised estimate remains near 12 even though comparable projects point to a much wider range.

Automation BiasThe team skips its normal estimate review because the model is treated as the authoritative estimator.

Risk score

Anchoring EffectA generated risk score becomes the numerical reference point around which later discussion moves.

Automation BiasReviewers ignore a missing input or obvious data-quality issue because the automated risk system still returned a recommendation.

Retrospective review protocol

Separate forecast quality, decision quality, and learning.

  1. Record the human judgment or range before automated advice when the decision is important enough to justify it.
  2. Separate the recommendation’s content from its source: what number or claim was supplied, and who or what supplied it?
  3. Compare against independent evidence, relevant base rates, and at least one alternative estimate.
  4. Ask whether changing the starting value changes the final number and whether changing the automated source changes the willingness to rely on it.
  5. Name only the pattern the evidence supports. An AI number can anchor a judgment without proving automation bias, and automation bias can occur without numerical anchoring.

Evidence

Reviewed sources behind this comparison.

  1. Fifty Years of Anchoring Effects: A Theoretical Reintegration and Meta-Analysis2026 · DOI 10.1287/mnsc.2023.03238
  2. Automation bias: a systematic review of frequency, effect mediators, and mitigators2011 · DOI 10.1136/amiajnl-2011-000089
  3. How was my performance? Exploring the role of anchoring bias in AI-assisted decision making2025 · DOI 10.1016/j.ijinfomgt.2025.102875

Comparison reviewed 2026-08-18. The individual bias pages contain their full evidence status and boundary conditions.