Accountability Capture: How Record-Keeping to Support AI Transparency and Accountability (Re)shapes Algorithmic Oversight

2025-10-06

AI Summary

This study examines the sociotechnical reconfiguration triggered by recordkeeping mechanisms in AI transparency and accountability practices, introducing and defining “accountability capture”—a phenomenon wherein formal documentation requirements retroactively reshape algorithmic governance, generating institutional tensions, employee resistance, and compliance conflicts. Drawing on an empirical survey of 100 AI practitioners, complemented by in-depth interviews and ethnographic observation, the research identifies a pervasive paradox: while recordkeeping aims to enhance traceability, it intensifies surveillance pressure, undermines data minimization principles, and weakens substantive accountability. The core contribution is a mechanistic account of how recordkeeping evolves from an instrumental practice into a structural constraint, offering a critical analytical framework for algorithmic governance. It highlights the profound tension between procedural justice—emphasizing adherence to documentation protocols—and substantive justice—centering on meaningful accountability outcomes—in technical governance regimes. (149 words)

Technology Category

• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Machine Learning: Ethics, Bias, and Fairness

Application Category

• Responsible Web: Algorithmic accountability and transparency on the web
• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

Abstract

Accountability regimes typically encourage record-keeping to enable the transparency that supports oversight, investigation, contestation, and redress. However, implementing such record-keeping can introduce considerations, risks, and consequences, which so far remain under-explored. This paper examines how record-keeping practices bring algorithmic systems within accountability regimes, providing a basis to observe and understand their effects. For this, we introduce, describe, and elaborate 'accountability capture' -- the re-configuration of socio-technical processes and the associated downstream effects relating to record-keeping for algorithmic accountability. Surveying 100 practitioners, we evidence and characterise record-keeping issues in practice, identifying their alignment with accountability capture. We further document widespread record-keeping practices, tensions between internal and external accountability requirements, and evidence of employee resistance to practices imposed through accountability capture. We discuss these and other effects for surveillance, privacy, and data protection, highlighting considerations for algorithmic accountability communities. In all, we show that implementing record-keeping to support transparency in algorithmic accountability regimes can itself bring wider implications -- an issue requiring greater attention from practitioners, researchers, and policymakers alike.

Problem

• Examining how record-keeping practices enable algorithmic system accountability
• Investigating unintended consequences of transparency-focused record-keeping implementation
• Analyzing tensions between internal and external accountability requirements

Innovation

• Introduces accountability capture concept for AI oversight
• Surveys practitioners to evidence record-keeping issues
• Documents tensions between internal external accountability requirements