Summary

Algorithms, and increasingly artificial intelligence (AI) algorithms, are relied upon by public authorities in high-stakes decision-making – where they are highly consequential for individuals’ lives. Within some jurisdictions, algorithmic outputs are being used as decisional support in bailing and recidivism decisions by judges; they have been deployed by school districts for teacher appraisals and to inform school reforms and individual firing decisions; by police to predict crime as well as in other areas as varied as healthcare, immigration or welfare. Their use has not been deprived of controversy. By virtue of their ‘black-box’ character and proprietary nature, AI algorithms can give rise to significant opacity challenges in terms of being able to interrogate the rationale behind decisions.

As artificial intelligence is permeating and shaping in fundamental ways our social and political worlds and our public institutions, issues of transparency and accountability are becoming pressing. A growing need to re-assess accountability in the era of automation is emerging: What accountability challenges do AI algorithmic systems bring with them, and how can we safeguard accountability in algorithmic decision-making?

The Speaker

Madalina Busuioc is Associate Professor at the Institute of Public Administration, Leiden University, where she leads, as principal investigator, a large European Research Council (ERC) grant for research on 'Regulatory Independence, Credibility and Accountability.'

Much of her work has focused on the study of public accountability and increasingly in this context, as public power is encoded algorithmically, her work on accountability has turned towards exploring the implications of AI systems for public accountability.

Her research has been published in JPART, Public Administration Review, Public Administration, Journal of European Public Policy, among others. She is field editor in Public Administration of the Journal of Public Policy and serves on the editorial board of the Journal of Public Administration Research and Theory (JPART).