The Bias Dilemma

The Ethics of Algorithmic Bias in Natural-Language Processing

Authors

DOI:

https://doi.org/10.5206/fpq/2022.3/4.14292

Keywords:

artificial intelligence, algorithms, bias

Abstract

Addressing biases in natural-language processing (NLP) systems presents an underappreciated ethical dilemma, which we think underlies recent debates about bias in NLP models. In brief, even if we could eliminate bias from language models or their outputs, we would thereby often withhold descriptively or ethically useful information, despite avoiding perpetuating or amplifying bias. Yet if we do not debias, we can perpetuate or amplify bias, even if we retain relevant descriptively or ethically useful information. Understanding this dilemma provides for a useful way of rethinking the ethics of algorithmic bias in NLP.

Downloads

Download data is not yet available.

Author Biographies

Oisín Deery, Macquarie University & York University

Oisín Deery is an assistant professor of philosophy at York University, Toronto, Canada, and a lecturer and ARC DECRA Fellow at Macquarie University in Sydney, Australia. He has previously held positions at Monash University, Florida State University, the University of Arizona, and the University of Montreal. Oisín’s work straddles the philosophy of mind and action, moral psychology, and metaphysics. His work has focused primarily on developing a naturalistic understanding of human agency and responsibility, but recently he has also been working on issues related to artificial intelligence and agency. Oisín has published widely in journals such as Philosophical Studies, the Australasian Journal of Philosophy, and Philosophical Psychology. He is the author of a monograph entitled Naturally Free Action, published by Oxford University Press in 2021.

Katherine Bailey, Shopify Inc.

Katherine Bailey has worked in a number of roles as a data scientist and software engineer. She is currently employed by Shopify Inc., in Toronto, Canada. Katherine has a longstanding interest in issues related to artificial intelligence and natural language processing. She has written on these issues in blog posts and in publications such as TechCrunch and has been invited to speak at various venues, including Google and the North American Association for Computational Linguistics.

Downloads

Published

2022-12-21

How to Cite

Deery, Oisín, and Katherine Bailey. 2022. “The Bias Dilemma: The Ethics of Algorithmic Bias in Natural-Language Processing”. Feminist Philosophy Quarterly 8 (3/4). https://doi.org/10.5206/fpq/2022.3/4.14292.

Issue

Vol. 8 No. 3/4 (2022): Feminism, Social Justice, and Artificial Intelligence

Section

Artificial Intelligence

License

Copyright (c) 2022 Oisín Deery, Katherine Bailey

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

The authors of work published in FPQ under the Creative Commons CC BY 4.0 License retain copyright to their work without restrictions and publication rights without restrictions. However, we request that authors include some sort of acknowledgement that the work was previously published in FPQ if part or all of a paper published in FPQ is used elsewhere.

Similar Articles

1 2 3 4 5 6 7 8 > >> 

You may also start an advanced similarity search for this article.