Evaluating the Impact of Explainable AI on Trust in AI-Assisted Code Review

2026-07-27

AI Summary

This study investigates how varying levels of explainability in AI-powered code review systems influence developers’ trust and actual adoption behaviors. Through a within-subjects user study involving 34 participants, the authors compare three conditions: no explanation, moderate feedback, and detailed explanations. Findings reveal that while detailed explanations significantly enhance perceived trust (mean = 3.99/5), they do not necessarily lead to higher adoption rates. In contrast, moderate feedback achieves the highest adoption rate at 89.22%. Notably, the level of explanation does not significantly affect code review time. These results highlight a decoupling between trust and behavioral adoption, offering empirical insights for designing effective explainable AI systems in software engineering contexts.

Abstract

Background: Large language models (LLMs) are increasingly used to automate code review, but the reasoning behind their decisions remains hard to understand. Developers struggle to assess the validity of LLM-generated reviews, making it difficult to gauge how much trust to place in them. The role of Explainable AI (XAI) in code review and its impact on trust remain underexplored. Objective: We study the influence of XAI on developer trust in AI-assisted code reviews. Method: We conducted a within-subjects user study with 34 participants, comparing three LLM-based code review systems with varying levels of XAI support: Condition A (detailed explanation and review feedback), Condition B (review feedback only), and Condition C (no explanations). Participants reviewed real-world code change requests alongside the AI-generated reviews. We measured trust perceptions, agreement with the AI recommendation, the reasoning given for each decision, and the time taken. Results: The level of explanation significantly influences both trust and agreement with AI recommendations, but in different ways. Full explanations (A) yield the highest perceived trust (M = 3.99/5) but not the highest agreement, whereas moderate explanations (B) achieve the highest agreement (89.22%). This could suggest that more explanation prompts developers to question AI recommendations more frequently. No explanations (C) results in the lowest trust and agreement. Explanation level did not significantly affect review time. The most commonly cited reasons for decisions were code readability and correctness. Conclusion: Incorporating XAI into code review significantly changes trust perceptions and agreement with AI recommendations. These results inform the design and evaluation of trustworthy AI-based code review systems, as well as studies on the human factors of AI-assisted software development.

Problem

• Explainable AI
• Trust
• Code Review
• Large Language Models
• Human-AI Interaction

Innovation

• Explainable AI
• Code Review
• Trust in AI
• Large Language Models
• Human-AI Interaction