Seeing Like a Toolkit: How Toolkits Envision the Work of AI Ethics

Abstract

Numerous toolkits have been developed to support ethical AI development. However, toolkits, like all tools, encode assumptions in their design about what work should be done and how. In this paper, we conduct a qualitative analysis of 27 AI ethics toolkits to critically examine how the work of ethics is imagined and how it is supported by these toolkits. Specifically, we examine the discourses toolkits rely on when talking about ethical issues, who they imagine should do the work of ethics, and how they envision the work practices involved in addressing ethics. Among the toolkits, we identify a mismatch between the imagined work of ethics and the support the toolkits provide for doing that work. In particular, we identify a lack of guidance around how to navigate labor, organizational, and institutional power dynamics as they relate to performing ethical work. We use these omissions to chart future work for researchers and designers of AI ethics toolkits.

Pre-print manuscript

References in corpus (8)

• Improving fairness in machine learning systems: What do industry practitioners need?
• Problem Formulation and Fairness
• Trust in Data Science: Collaboration, Translation, and Accountability in Corporate Data Science Projects
• Everyday algorithm auditing: Understanding the power of everyday users in surfacing harmful algorithmic behaviors
• Exploring How Machine Learning Practitioners (Try To) Use Fairness Toolkits
• Principles to Practices for Responsible AI: Closing the Gap
• Value Cards: An Educational Toolkit for Teaching Social Impacts of Machine Learning through Deliberation
• Soliciting Stakeholders' Fairness Notions in Child Maltreatment Predictive Systems

Cited by in corpus (18)

• Farsight: Fostering Responsible AI Awareness During AI Application Prototyping
• The Situate AI Guidebook: Co-Designing a Toolkit to Support Multi-Stakeholder Early-stage Deliberations Around Public Sector AI Proposals
• Inherent Limitations of AI Fairness
• A Scoping Study of Evaluation Practices for Responsible AI Tools: Steps Towards Effectiveness Evaluations
• Towards AI Accountability Infrastructure: Gaps and Opportunities in AI Audit Tooling
• The Value-Sensitive Conversational Agent Co-Design Framework
• Ethics Pathways: A Design Activity for Reflecting on Ethics Engagement in HCI Research
• Legacy Procurement Practices Shape How U.S. Cities Govern AI: Understanding Government Employees' Practices, Challenges, and Needs
• "Come to us first": Centering Community Organizations in Artificial Intelligence for Social Good Partnerships
• Making the Right Thing: Bridging HCI and Responsible AI in Early-Stage AI Concept Selection
• Thoughtful Adoption of NLP for Civic Participation: Understanding Differences Among Policymakers
• Supporting Industry Computing Researchers in Assessing, Articulating, and Addressing the Potential Negative Societal Impact of Their Work
• "Why do we do this?": Moral Stress and the Affective Experience of Ethics in Practice
• Towards Creating Infrastructures for Values and Ethics Work in the Production of Software Technologies
• "Near Data" and "Far Data" for Urban Sustainability: How Do Community Advocates Envision Data Intermediaries?
• Identifying the Barriers to Human-Centered Design in the Workplace: Perspectives from UX Professionals
• Talking About the Assumption in the Room
• Compliant But Unsatisfactory: The Gap Between Auditing Standards and Practices for Probabilistic Genotyping Software