Algorithm aversion: People erroneously avoid algorithms after seeing them err.

2014/11/17 by Berkeley J. Dietvorst, Joseph P. Simmons, Cade Massey · 80 citations
Decision Sciences · Social Sciences · #Decision-Making and Behavioral Economics #Forecasting Techniques and Applications #Innovation, Sustainability, Human-Machine Systems

paper · doi:10.1037/xge0000033

openalex publication_date 2014/11/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/31

Abstract

Research shows that evidence-based algorithms more accurately predict the future than do human forecasters. Yet when forecasters are deciding whether to use a human forecaster or a statistical algorithm, they often choose the human forecaster. This phenomenon, which we call algorithm aversion, is costly, and it is important to understand its causes. We show that people are especially averse to algorithmic forecasters after seeing them perform, even when they see them outperform a human forecaster. This is because people more quickly lose confidence in algorithmic than human forecasters after seeing them make the same mistake. In 5 studies, participants either saw an algorithm make forecasts, a human make forecasts, both, or neither. They then decided whether to tie their incentives to the future predictions of the algorithm or the human. Participants who saw the algorithm perform were less confident in it, and less likely to choose it over an inferior human forecaster. This was true even among those who saw the algorithm outperform the human.

Citations

• The robust beauty of improper linear models in decision making.
• Factors influencing the use of a decision rule in a probabilistic task
• The relative influence of advice from human experts and statistical methods on forecast adjustments
• Clinical versus mechanical prediction: A meta-analysis.
• Stubborn Reliance on Intuition and Subjectivity in Employee Selection
• The trouble with overconfidence.

Cited by

• Human-in-the-Loop User Feedback Affects Perceived Accuracy and Trust, but Task Subjectivity Matters
• FIFA World Cup 2026 as a Contamination-Free Benchmark for LLM Forecasting Agents: Four Models, a Bookmaker, and 104 Matches
• When Not to Automate: A Formal Protocol for Human Preservation in AI-Optimized Organizations
• Align AI to Dynamic Human-AI Workflows
• Can AR Embedded Visualizations Foster Appropriate Reliance on AI in Spatial Decision-Making? A Comparative Study of AR X-Ray vs. 2D Minimap
• Ads in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest
• Reliability of LLMs as medical assistants for the general public: a randomized preregistered study
• Everyone prefers human writers, including AI
• AI Self-preferencing in Algorithmic Hiring: Empirical Evidence and Insights
• Rethinking How We Theorize AI in Organization and Management: A Problematizing Review of Rationality and Anthropomorphism
• Large Language Models are Near-Optimal Decision-Makers with a Non-Human Learning Behavior
• Beyond Adoption Intention How Trust in Augmented Analytics Relates to Perceived Decision Quality Among Non-Technical BI Users
• Verification Without Distrust: Reframing User-Side Oversight as Routine Epistemic Governance in Everyday Human-Chatbot Interaction
• Mental Models of Autonomy and Sentience Shape Reactions to AI
• When Medical AI Explanations Help and When They Harm
• Simulating Life Paths with Digital Twins: AI-Generated Future Selves Influence Decision-Making and Expand Human Choice
• LLM-Generated Ads: From Personalization Parity to Persuasion Superiority
• Humans incorrectly reject confident accusatory AI judgments
• Optimal Comprehensible Targeting
• Design principles for text-to-image generative artificial intelligence creativity support tools for visual design
• Revealing AI Reasoning Increases Trust but Crowds Out Unique Human Knowledge
• Opportunity Versus Threat Appraisals of AI Aids: The Effect of Appraisal Type on Decision Makers' Effort and Compliance When Using Powerful AI Aids
• The Innate Economic Preferences of Language Models
• A meta-analysis on reactions to algorithmic decision-making in human resource management
• Understanding Reader Perception Shifts upon Disclosure of AI Authorship
• Catch Me If You Can: The Dynamic Nature of Bias in Machine Learning Applications
• When AI Gives Advice: Evaluating AI and Human Responses to Online Advice-Seeking for Well-Being
• Human-Robo-advisor collaboration in decision-making: Evidence from a multiphase mixed methods experimental study
• The social consequences of AI delegation
• PriorWeaver: Prior Elicitation via Iterative Dataset Construction
• Lay belief about AI and its decision-making
• Dialogues with AI Reduce Beliefs in Misinformation but Build No Lasting Discernment Skills
• Does AI Coaching Prepare us for Workplace Negotiations?
• Position: Human Factors Reshape Adversarial Analysis in Human-AI Decision-Making Systems
• Algorithmic Governance: Experimental Evidence on Citizens' and Public Administrators' Legitimacy Perceptions of Automated Decision‐Making
• A co-evolving agentic AI system for medical imaging analysis
• Support for AI in Public Administration: A Comparative Study of the United Kingdom and Japan
• “It Became My Buddy, But I’m Not Afraid to Disagree”: A Multi-Session Study of UX Evaluators Collaborating with Conversational AI Assistants
• Perceived Fairness of Human Managers Compared with Artificial Intelligence in Employee Performance Evaluation
• AI Knows Best? The Paradox of Expertise, AI-Reliance, and Performance in Educational Tutoring Decision-Making Tasks
• A Framework for Generating Artificial Datasets to Validate Absolute and Relative Position Concepts
• Fairness-Aware and Interpretable Policy Learning
• Exploring Conversational Design Choices in LLMs for Pedagogical Purposes: Socratic and Narrative Approaches for Improving Instructor's Teaching Practice
• 'Alexa, Do You Know Anything?' The Impact of an Intelligent Assistant on Team Interactions and Creative Performance Under Time Scarcity
• Human or Robot? Evidence from Last-Mile Delivery Service
• Proactive AI Adoption can be Threatening: When Help Backfires
• Bias in the Loop: How Humans Evaluate AI-Generated Suggestions
• Measuring and mitigating overreliance to build human-compatible AI
• Beyond ATE: Multi-Criteria Design for A/B Testing
• Would I regret being different? The influence of social norms on attitudes toward AI usage
• What influences algorithmic decision-making? A systematic literature review on algorithm aversion
• FinTech and consumers: a systematic review and integrative framework
• Overcoming medical overuse with AI assistance: An experimental investigation
• Algorithm appreciation: People prefer algorithmic to human judgment
• Making sense of recommendations
• Designing, Implementing, and Evaluating AI Explanations: A Scoping Review of Explainable AI Frameworks
• AI can help humans find common ground in democratic deliberation
• The role of trust and algorithms in consumers’ front-of-pack labels acceptance: a cross-country investigation
• When do you trust AI? The effect of number presentation detail on consumer trust and acceptance of AI recommendations
• An Initial Model of Trust in Chatbots for Customer Service—Findings from a Questionnaire Study
• Evidence of a social evaluation penalty for using AI
• The flaws of policies requiring human oversight of government algorithms
• Beyond Predictions: A Study of AI Strength and Weakness Transparency Communication on Human-AI Collaboration
• Heterogeneous preferences and asymmetric insights for AI use among welfare claimants and non-claimants
• Overcoming Algorithm Aversion with Transparency: Can Transparent Predictions Change User Behavior?
• Why Do Decision Makers (Not) Use AI? A Cross-Domain Analysis of Factors Impacting AI Adoption
• Understanding Consumer Preferences for Explanations Generated by XAI Algorithms
• Human vs. Algorithmic Auditors: The Impact of Entity Type and Ambiguity on Human Dishonesty
• Understanding how users may work around algorithmic bias
• The Endless Tuning. An Artificial Intelligence Design To Avoid Human Replacement and Trace Back Responsibilities
• Artificial intelligence and the image war: does exposing AI-generated images as fake limit the images’ influence on perceptions about wars and conflicts?
• Machine Learning: An Applied Econometric Approach
• Nine Potential Pitfalls when Designing Human-AI Co-Creative Systems
• AI aversion? Effects of author disclosure on young people’s perceptions of mental health advice
• Made With AI: Consumer Engagement With Social Media Containing AI Disclosures
• Attri-Net: A Globally and Locally Inherently Interpretable Model for Multi-Label Classification Using Class-Specific Counterfactuals
• Beyond algorithm aversion: How users trade off decisional agency and optimal outcomes when choosing between algorithms and human decision-makers
• English language teachers' perspectives on digital trust in AI‐supported education
• Decision authority and the returns to algorithms
• ArchiveGPT: A human-centered evaluation of using a vision language model for image cataloguing

Related

• Overcoming Algorithm Aversion: People Will Use Imperfect Algorithms If They Can (Even Slightly) Modify Them
• A systematic review of algorithm aversion in augmented decision making
• What influences algorithmic decision-making? A systematic literature review on algorithm aversion
• Reputational Algorithm Aversion
• An Integrative Perspective on Algorithm Aversion and Appreciation in Decision-Making