10 things you should know about algorithmic fairness

Woodruff, Allison

doi:10.1145/3328489

Cited by 10 publications

(4 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These findings reinforce and extend previous arguments that broad requirements to provide explanations are unlikely to meet the full spectrum of public needs [5,69,73,86]. Our research also underscores that explanation per se is not necessarily the end goal [58,77,100], but rather there are goals underlying requests for explanation that can often be met via other mechanisms that provide accountability, contestability, appeal, advancement of human knowledge, or other qualities.…”

Section: Design Choices Related To Explainability Should Account For ...supporting

confidence: 88%

“…More fundamentally, there is a growing recognition that transparency, explainability, and interpretability are often intended as a means to an end, for example laying the groundwork to challenge a decision or to ensure fairness, yet they provide inadequate support for such underlying objectives [58,77,100]. As a result, a number of approaches have been articulated in recent years that hew more closely to direct needs of users and society.…”

Section: Accountability Transparency Explainability and Interpretabilitymentioning

confidence: 99%

See 1 more Smart Citation

"A cold, technical decision-maker": Can AI provide explainability, negotiability, and humanity?

Woodruff¹,

Asare²,

Jameson³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Algorithmic systems are increasingly deployed to make decisions in many areas of people's lives. The shift from human to algorithmic decision-making has been accompanied by concern about potentially opaque decisions that are not aligned with social values, as well as proposed remedies such as explainability. We present results of a qualitative study of algorithmic decision-making, comprised of five workshops conducted with a total of 60 participants in Finland, Germany, the United Kingdom, and the United States. We invited participants to reason about decision-making qualities such as explainability and accuracy in a variety of domains. Participants viewed AI as a decision-maker that follows rigid criteria and performs mechanical tasks well, but is largely incapable of subjective or morally complex judgments. We discuss participants' consideration of humanity in decision-making, and introduce the concept of 'negotiability', the ability to go beyond formal criteria and work flexibly around the system. CCS Concepts: • Social and professional topics → Computing / technology policy; • Human-centered computing → Human computer interaction (HCI); • Computing methodologies → Artificial intelligence.

show abstract

Section: Design Choices Related To Explainability Should Account For ...supporting

confidence: 88%

Section: Accountability Transparency Explainability and Interpretabilitymentioning

confidence: 99%

"A cold, technical decision-maker": Can AI provide explainability, negotiability, and humanity?

Woodruff¹,

Asare²,

Jameson³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The idea of contestability in technology systems can be traced back to early expert and mixedinitiative systems where experts negotiate with or correct the system to optimize its output [42,87,106]. More recently, researchers have returned to this idea, arguing that the algorithmic experience can be improved by allowing users more of a voice in how decision are made [115,119].…”

Section: Designing For Contestabilitymentioning

confidence: 99%

Contestability For Content Moderation

Vaccaro

Xiao

Hamilton

et al. 2021

Proc. ACM Hum.-Comput. Interact.

View full text Add to dashboard Cite

Content moderation systems for social media have had numerous issues of bias, in terms of race, gender, and ability among many others. One proposal for addressing such issues in automated decision making is by designing for contestability, whereby users can shape and influence how decisions are made. In this study, we conduct a series of participatory design workshops with participants from communities that have experienced problems with social media content moderation in the past. Together with participants, we explore the idea of designing for contestability in content moderation and find that users' designs suggest three fruitful, practical avenues: adding representation, improving communication, and designing with compassion. We conclude with design recommendations drawn from participants' proposals, and reflect on the challenges that remain.CCS Concepts: • Human-centered computing → Human computer interaction (HCI); Social media.

show abstract

“…We suggest that one way fairness has in the past been fruitfully deployed in the context of administering justice (i.e., making legal decisions) is in “access to justice” studies. We approach algorithmisation and algorithmic fairness as a wicked complex societal problem (Selbst et al, 2019 ; Woodruff, 2019 ), in need of interdisciplinary research. For us, algorithmic fairness is broader than the quest to reduce bias in algorithms.…”

Section: Introductionmentioning

confidence: 99%

Research agenda for algorithmic fairness studies: Access to justice lessons for interdisciplinary research

Kontiainen

Koulu

Sankari

2022

Front. Artif. Intell.

View full text Add to dashboard Cite

Access to justice is one of the fundamental legitimating principles underlying all modern Western legal systems, yet its role in critical algorithm studies remains underdeveloped. In historical and methodological terms, the access to justice movement showcased multi- and interdisciplinary research on legal phenomena. We argue that interdisciplinary research on AI ethics and regulation, datafication of society, and algorithmic governance could benefit from adopting access to justice as a vantage point for bridging the different approaches in the context of administering justice. To this end, we explore technological, legal, and societal intersections to demonstrate how law, social sciences, and algorithm studies could benefit from a historically more informed and holistic approach facilitating more “cost-effective” interdisciplinary research collaboration. Such approach could assist the substantive study of algorithmic fairness to contribute actionable systemic solutions on what we perceive as systemic challenges. We propose utilizing access to justice as a boundary object for interdisciplinary dialogue over algorithmic fairness while respecting the epistemic diversity of disciplines.

show abstract

10 things you should know about algorithmic fairness

Cited by 10 publications

References 5 publications

"A cold, technical decision-maker": Can AI provide explainability, negotiability, and humanity?

"A cold, technical decision-maker": Can AI provide explainability, negotiability, and humanity?

Contestability For Content Moderation

Research agenda for algorithmic fairness studies: Access to justice lessons for interdisciplinary research

Contact Info

Product

Resources

About