Algorithmic Impact Assessments and Accountability: The Co-construction of Impacts. Metcalf, J., Moss, E., Watkins, E. A., Singh, R., & Elish, M. C. In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, of FAccT '21, pages 735–746, New York, NY, USA, March, 2021. Association for Computing Machinery. ![Algorithmic Impact Assessments and Accountability: The Co-construction of Impacts [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Algorithmic impact assessments (AIAs) are an emergent form of accountability for organizations that build and deploy automated decision-support systems. They 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 actors to identify and ameliorate harms experienced because of a policy decision or system. Every domain has different expectations and norms around what constitutes impacts and harms, how potential harms are rendered as impacts of a particular undertaking, who is responsible for conducting such assessments, and who has the authority to act on them 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. As impact assessments become a commonplace process for evaluating harms, the FAccT community, in its efforts to address this challenge, should A) understand impacts as objects that are co-constructed accountability relationships, 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 building robust accountability relationships.
@inproceedings{metcalf_algorithmic_2021,
address = {New York, NY, USA},
series = {{FAccT} '21},
title = {Algorithmic {Impact} {Assessments} and {Accountability}: {The} {Co}-construction of {Impacts}},
isbn = {978-1-4503-8309-7},
shorttitle = {Algorithmic {Impact} {Assessments} and {Accountability}},
url = {https://doi.org/10.1145/3442188.3445935},
doi = {10.1145/3442188.3445935},
abstract = {Algorithmic impact assessments (AIAs) are an emergent form of accountability for organizations that build and deploy automated decision-support systems. They 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 actors to identify and ameliorate harms experienced because of a policy decision or system. Every domain has different expectations and norms around what constitutes impacts and harms, how potential harms are rendered as impacts of a particular undertaking, who is responsible for conducting such assessments, and who has the authority to act on them 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. As impact assessments become a commonplace process for evaluating harms, the FAccT community, in its efforts to address this challenge, should A) understand impacts as objects that are co-constructed accountability relationships, 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 building robust accountability relationships.},
urldate = {2021-03-12},
booktitle = {Proceedings of the 2021 {ACM} {Conference} on {Fairness}, {Accountability}, and {Transparency}},
publisher = {Association for Computing Machinery},
author = {Metcalf, Jacob and Moss, Emanuel and Watkins, Elizabeth Anne and Singh, Ranjit and Elish, Madeleine Clare},
month = mar,
year = {2021},
keywords = {accountability, algorithmic impact assessment, governance, harm, impact},
pages = {735--746},
}
Downloads: 0
{"_id":"YZPhSWbsvDBFB9cxR","bibbaseid":"metcalf-moss-watkins-singh-elish-algorithmicimpactassessmentsandaccountabilitythecoconstructionofimpacts-2021","author_short":["Metcalf, J.","Moss, E.","Watkins, E. A.","Singh, R.","Elish, M. C."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"New York, NY, USA","series":"FAccT '21","title":"Algorithmic Impact Assessments and Accountability: The Co-construction of Impacts","isbn":"978-1-4503-8309-7","shorttitle":"Algorithmic Impact Assessments and Accountability","url":"https://doi.org/10.1145/3442188.3445935","doi":"10.1145/3442188.3445935","abstract":"Algorithmic impact assessments (AIAs) are an emergent form of accountability for organizations that build and deploy automated decision-support systems. They 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 actors to identify and ameliorate harms experienced because of a policy decision or system. Every domain has different expectations and norms around what constitutes impacts and harms, how potential harms are rendered as impacts of a particular undertaking, who is responsible for conducting such assessments, and who has the authority to act on them 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. As impact assessments become a commonplace process for evaluating harms, the FAccT community, in its efforts to address this challenge, should A) understand impacts as objects that are co-constructed accountability relationships, 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 building robust accountability relationships.","urldate":"2021-03-12","booktitle":"Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency","publisher":"Association for Computing Machinery","author":[{"propositions":[],"lastnames":["Metcalf"],"firstnames":["Jacob"],"suffixes":[]},{"propositions":[],"lastnames":["Moss"],"firstnames":["Emanuel"],"suffixes":[]},{"propositions":[],"lastnames":["Watkins"],"firstnames":["Elizabeth","Anne"],"suffixes":[]},{"propositions":[],"lastnames":["Singh"],"firstnames":["Ranjit"],"suffixes":[]},{"propositions":[],"lastnames":["Elish"],"firstnames":["Madeleine","Clare"],"suffixes":[]}],"month":"March","year":"2021","keywords":"accountability, algorithmic impact assessment, governance, harm, impact","pages":"735–746","bibtex":"@inproceedings{metcalf_algorithmic_2021,\n\taddress = {New York, NY, USA},\n\tseries = {{FAccT} '21},\n\ttitle = {Algorithmic {Impact} {Assessments} and {Accountability}: {The} {Co}-construction of {Impacts}},\n\tisbn = {978-1-4503-8309-7},\n\tshorttitle = {Algorithmic {Impact} {Assessments} and {Accountability}},\n\turl = {https://doi.org/10.1145/3442188.3445935},\n\tdoi = {10.1145/3442188.3445935},\n\tabstract = {Algorithmic impact assessments (AIAs) are an emergent form of accountability for organizations that build and deploy automated decision-support systems. They 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 actors to identify and ameliorate harms experienced because of a policy decision or system. Every domain has different expectations and norms around what constitutes impacts and harms, how potential harms are rendered as impacts of a particular undertaking, who is responsible for conducting such assessments, and who has the authority to act on them 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. As impact assessments become a commonplace process for evaluating harms, the FAccT community, in its efforts to address this challenge, should A) understand impacts as objects that are co-constructed accountability relationships, 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 building robust accountability relationships.},\n\turldate = {2021-03-12},\n\tbooktitle = {Proceedings of the 2021 {ACM} {Conference} on {Fairness}, {Accountability}, and {Transparency}},\n\tpublisher = {Association for Computing Machinery},\n\tauthor = {Metcalf, Jacob and Moss, Emanuel and Watkins, Elizabeth Anne and Singh, Ranjit and Elish, Madeleine Clare},\n\tmonth = mar,\n\tyear = {2021},\n\tkeywords = {accountability, algorithmic impact assessment, governance, harm, impact},\n\tpages = {735--746},\n}\n\n","author_short":["Metcalf, J.","Moss, E.","Watkins, E. A.","Singh, R.","Elish, M. C."],"key":"metcalf_algorithmic_2021","id":"metcalf_algorithmic_2021","bibbaseid":"metcalf-moss-watkins-singh-elish-algorithmicimpactassessmentsandaccountabilitythecoconstructionofimpacts-2021","role":"author","urls":{"Paper":"https://doi.org/10.1145/3442188.3445935"},"keyword":["accountability","algorithmic impact assessment","governance","harm","impact"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://api.zotero.org/users/1922542/collections/4USSPCM8/items?key=2db6Wad6ty95LBsQ3vWXMPvG&format=bibtex&limit=100","dataSources":["GRN6ADMdfx7xrpxLd"],"keywords":["accountability","algorithmic impact assessment","governance","harm","impact"],"search_terms":["algorithmic","impact","assessments","accountability","construction","impacts","metcalf","moss","watkins","singh","elish"],"title":"Algorithmic Impact Assessments and Accountability: The Co-construction of Impacts","year":2021}