Algorithmic Impact Assessments and Accountability: The Co-construction of Impacts

Jacob Metcalf, Emanuel Moss, Elizabeth Anne Watkins, Ranjit Singh, and MadeleineClare Elish. 2021. Algorithmic Impact Assessments and Accountability: TheCo-construction of Impacts. InACM Conference on Fairness, Accountability,and Transparency (FAccT ’21), March 3–10, 2021, Virtual Event, Canada.ACM,

19 Pages Posted: 12 Feb 2021

See all articles by Jacob Metcalf

Jacob Metcalf

Data & Society Research Institute

Emanuel Moss

Intel Labs

Elizabeth Anne Watkins

Princeton University Center for Information Technology Policy; Data & Society Research Institute

Ranjit Singh

Data & Society Research Institute

Madeleine Clare Elish

Google Inc.; University of Oxford - Oxford Internet Institute

Date Written: September 29, 2020

Abstract

Algorithmic impact assessments (AIAs) are an emergent form of accountability for entities that build and deploy automated decision-support systems. These are modeled after impact assessments in other domains. Our study of the history of impact assessments shows that "impacts" are an evaluative construct that enable institutions to identify and ameliorate harms experienced because of a policy decision or system. Every domain has different expectations and norms about what constitutes impacts and harms, how potential harms are rendered as the impacts of a particular undertaking, who is responsible for conducting that assessment, and who has the authority to act on the impact assessment to demand changes to that undertaking. By examining proposals for AIAs in relation to other domains, we find that there is a distinct risk of constructing algorithmic impacts as organizationally understandable metrics that are nonetheless inappropriately distant from the harms experienced by people, and which fall short of building the relationships required for effective accountability. To address this challenge of algorithmic accountability, and as impact assessments become a commonplace process for evaluating harms, the FAccT community should A) understand impacts as objects constructed for evaluative purposes, B) attempt to construct impacts as close as possible to actual harms, and C) recognize that accountability governance requires the input of various types of expertise and affected communities. We conclude with lessons for assembling cross-expertise consensus for the co-construction of impacts and to build robust accountability relationships.

Keywords: algorithmic impact assessment, impact, harm, accountability, governance

Suggested Citation

Metcalf, Jacob and Moss, Emanuel and Watkins, Elizabeth and Watkins, Elizabeth and Singh, Ranjit and Elish, Madeleine Clare, Algorithmic Impact Assessments and Accountability: The Co-construction of Impacts (September 29, 2020). Jacob Metcalf, Emanuel Moss, Elizabeth Anne Watkins, Ranjit Singh, and MadeleineClare Elish. 2021. Algorithmic Impact Assessments and Accountability: TheCo-construction of Impacts. InACM Conference on Fairness, Accountability,and Transparency (FAccT ’21), March 3–10, 2021, Virtual Event, Canada.ACM,, Available at SSRN: https://ssrn.com/abstract=3736261

Jacob Metcalf

Data & Society Research Institute ( email )

36 West 20th Street
11th Floor
New York,, NY 10011
United States

Emanuel Moss (Contact Author)

Intel Labs ( email )

2200 Mission College Blvd.
Santa Clara, CA 95054-1549
United States

Elizabeth Watkins

Princeton University Center for Information Technology Policy ( email )

C231A E-Quad
Olden Street
Princeton, NJ 08540
United States

HOME PAGE: http://https://citp.princeton.edu/citp-people/watkins/

Data & Society Research Institute ( email )

36 West 20th Street
11th Floor
New York,, NY 10011
United States

Ranjit Singh

Data & Society Research Institute ( email )

36 West 20th Street
11th Floor
New York,, NY 10011
United States

Madeleine Clare Elish

Google Inc. ( email )

1600 Amphitheatre Parkway
Second Floor
Mountain View, CA 94043
United States

University of Oxford - Oxford Internet Institute ( email )

1 St. Giles
University of Oxford
Oxford OX1 3PG Oxfordshire, Oxfordshire OX1 3JS
United Kingdom

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
762
Abstract Views
3,624
Rank
60,763
PlumX Metrics