State of the Evidence: Algorithmic Transparency

Computational systems are everywhere mediating the fabric of contemporary social life, from public transportation and welfare provision to content moderation and criminal prosecution. Algorithmic decision-making (ADM) systems are increasingly intervening in government and business processes of all kinds. Search engines and social media platforms are deeply imbued by algorithmic systems with the power to rank, classify, moderate, or shape information and our social relations. Yet their decision-making processes are frequently opaque and inscrutable for citizens or even their developers.

This literature review by Matías Valderrama, María Paz Hermosilla, and Romina Garrido focuses on transparencyAccording to OGP’s Articles of Governance, transparency occurs when “government-held information (including on activities and decisions) is open, comprehensive, timely, freely available to the pub... More in a broader sense and seeks to bring more clarity to the concept of algorithmic transparency by identifying some of the main mechanisms that have been proposed in the literature to promote it. The main objective of this review is to examine the available evidence on the outcomes or results of such mechanisms. In this way, we want to contribute with an early stock-taking exercise of the research on impacts of algorithmic transparency in the public sector. This work is primarily intended for decision-makers to facilitate more precise and targeted policy discussions. Additionally, anyone who develops or works with algorithmic systems and wants to know the mechanisms under discussion to examine, evaluate, and make that system more transparent may find it useful.