AI systems that skip advice affect users differently by expertise
The Differential Effects of AI Abstention on Advice Acceptance
Human-Computer Interaction
Summary
Sometimes AI systems decide not to give advice if their suggestions might not be good. The authors studied how people respond to such AI systems that give advice selectively versus those that always give advice. They found that when AI withholds advice, people who feel like experts are more likely to trust and accept the AI’s advice, while people who feel less expert trust it less. This is because expert users find it easier to balance advice with their own knowledge when the AI abstains from bad suggestions. This approach also helps expert users keep trusting the AI advice better over many decisions.
What this means in practice
- •For financial advisors: Design AI tools that selectively provide advice to improve decision support for clients with different expertise levels.
- •For customer support managers: Implement AI systems that withhold recommendations occasionally to maintain user trust, especially for experienced users in repeated interactions.
Authors
Nicholas Wolczynski, Yotam Liel, Maytal Saar-Tsechansky, Lior Zalmanson
Abstract
AI advising systems are typically designed to always provide a response. Recent work has focused on developing systems that withhold advice in some instances, such as when advice quality is questionable, to improve the overall quality of guidance when it is given; a design paradigm we refer to as AI abstention. While substantial work has focused on optimizing when models should abstain, little is known about how users respond to systems that abstain. In a preregistered experiment, participants made repeated financial decisions, assisted by one of two advising regimes: an AI that advises selectively or one that advises on every decision. We find that the same AI abstention regime produces differential effects depending on the user: it increases advice acceptance among users with high perceived expertise and tends to decrease it among users with low perceived expertise. This differential effect is accounted for by the cognitive burden users face when repeatedly reconciling AI recommendations with their own judgment; AI abstention appears to relieve this burden among users with high perceived expertise, and in doing so increases their receptivity to advice. Further, AI abstention mitigates the decline in advice acceptance observed among expert users over repeated interactions. Exploratory evidence also suggests it increases appropriate acceptance where advice would improve the decision, increasing the downstream accuracy of perceived experts. The AI abstention regime is thus a first-order design property of AI advisory systems, and one that must be calibrated to the users a system serves.