AI2030 Responsible AI Tools Marketplace

Key Takeaways

Interactive explanations build more trust than static ones: The study found that when AI systems provided interactive explanations (such as counterfactuals showing "what-if" scenarios), users reported higher confidence and engagement compared to basic feature importance explanations alone.
Clarity and relevance matter more than complexity: The research demonstrates that how well an explanation matches what users actually need to understand—rather than technical sophistication—is the key factor determining whether people trust an AI system's decisions.
Explainability design has measurable real-world impact: By testing explanations in a realistic loan approval scenario, the study provides concrete evidence that the way companies choose to explain AI decisions directly affects whether users will trust those systems in high-stakes domains like finance, healthcare, and law.
One-size-fits-all explanations are insufficient: The comparison of different explanation types suggests that organizations deploying AI in critical domains should consider tailoring explanations to user needs rather than relying on a single explanation method.

Abstract

Large-scale AI models such as GPT-4 have accelerated the deployment of artificial intelligence across critical domains including law, healthcare, and finance, raising urgent questions about trust and transparency. This study investigates the relationship between explainability and user trust in AI systems through a quantitative experimental design. Using an interactive, web-based loan approval simulation, we compare how different types of explanations, ranging from basic feature importance to interactive counterfactuals influence perceived trust. Results suggest that interactivity enhances both user engagement and confidence, and that the clarity and relevance of explanations are key determinants of trust. These findings contribute empirical evidence to the growing field of human-centered explainable AI, highlighting measurable effects of explainability design on user perception

8 pages, 3 figures, 2 appendices. Quantitative user study on AI explainability and trust. Preprint, 2025

Key Takeaways

Interactive explanations build more trust than static ones: The study found that when AI systems provided interactive explanations (such as counterfactuals showing "what-if" scenarios), users reported higher confidence and engagement compared to basic feature importance explanations alone.
Clarity and relevance matter more than complexity: The research demonstrates that how well an explanation matches what users actually need to understand—rather than technical sophistication—is the key factor determining whether people trust an AI system's decisions.
Explainability design has measurable real-world impact: By testing explanations in a realistic loan approval scenario, the study provides concrete evidence that the way companies choose to explain AI decisions directly affects whether users will trust those systems in high-stakes domains like finance, healthcare, and law.
One-size-fits-all explanations are insufficient: The comparison of different explanation types suggests that organizations deploying AI in critical domains should consider tailoring explanations to user needs rather than relying on a single explanation method.

Abstract

8 pages, 3 figures, 2 appendices. Quantitative user study on AI explainability and trust. Preprint, 2025

Preliminary Quantitative Study on Explainability and Trust in AI Systems

Key Takeaways

Abstract

Paper Info

Responsible AI Topics

Links

Preliminary Quantitative Study on Explainability and Trust in AI Systems

Key Takeaways

Abstract

Paper Info

Responsible AI Topics

Links