WeAudit: Scaffolding User Auditors and AI Practitioners in Auditing Generative AI

2025-01-02

AI Summary

Ordinary users struggle to effectively participate in auditing generative AI content, and their feedback is rarely incorporated by industry practitioners. Method: We propose WeAudit—the first structured framework supporting user-centered participatory AI auditing—integrating reflective interaction design and a cross-role feedback loop to enable both individual and collaborative auditing, thereby aligning user-identified issues, problem representation, and developer responses. Grounded in human-AI collaboration principles, WeAudit was validated through qualitative formal research, iterative prototyping, a three-week in-the-wild user study, and in-depth interviews with AI practitioners. Contribution/Results: WeAudit significantly enhances users’ ability to detect potential AI harms and generates audit reports that are both engineering-interpretable and actionable. These outputs received strong endorsement from AI practitioners, demonstrating the framework’s practical viability and impact on closing the gap between end-user insights and industrial AI development practices.

Technology Category

• Humans and AI: User Experience and Usability
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

There has been growing interest from both practitioners and researchers in engaging end users in AI auditing, to draw upon users' unique knowledge and lived experiences. However, we know little about how to effectively scaffold end users in auditing in ways that can generate actionable insights for AI practitioners. Through formative studies with both users and AI practitioners, we first identified a set of design goals to support user-engaged AI auditing. We then developed WeAudit, a workflow and system that supports end users in auditing AI both individually and collectively. We evaluated WeAudit through a three-week user study with user auditors and interviews with industry Generative AI practitioners. Our findings offer insights into how WeAudit supports users in noticing and reflecting upon potential AI harms and in articulating their findings in ways that industry practitioners can act upon. Based on our observations and feedback from both users and practitioners, we identify several opportunities to better support user engagement in AI auditing processes. We discuss implications for future research to support effective and responsible user engagement in AI auditing and red-teaming.

Problem

• User Engagement
• AI Content Verification
• Expert Feedback

Innovation

• WeAudit
• AI Auditing
• User Engagement