Understanding how users may work around algorithmic bias
Bibliographic Data
| ID | 20398367 |
|---|---|
| Authors | Hannah Overbye-Thompson (0000-0001-5235-9486, University of California, Santa Barbara), R E Rice (0000-0002-6374-7860, University of California, Santa Barbara, corresponding author) |
| Year | 2026 |
| Volume | 41 |
| Issue | 3 |
| Pages | 2385-2399 |
| Publication date | 2026-03-01 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | AI & Society (JOURNAL) |
| Journal identifiers | ISSN: 0951-5666 • E-ISSN: 1435-5655 |
| Publisher | Springer Science and Business Media LLC (PUBLISHER) |
| DOI | 10.1007/s00146-025-02498-1 |
| OpenAlex | W4412676577 |
| Language | EN |
| Citations received | 1 |
| References cited | 71 |
Algorithms increasingly mediate critical aspects of daily life across healthcare, hiring, and social media, shaping user experiences through automated decision-making processes. Yet, algorithmic bias, the systematic disadvantaging of certain groups through automated systems, has been widely documented across a variety of algorithms. Thus this study addresses the gap in understanding how users may respond to four epistemic categories of algorithm bias, depending on whether it exists or not, and is perceived or not. We apply the information systems concept of workarounds to characterize potential user responses to these categories of algorithmic bias. Then, we apply the Human–AI Interaction Theory of Interactive Media Effects to understand how users may detect bias through cue routes and develop workaround strategies through action routes. Our theoretical framework proposes how users' detection and workarounds may vary based on the four categories of bias. Understanding these adaptive strategies provides crucial insights for developing inclusive technologies and fostering algorithmic literacy, ultimately enhancing the ongoing negotiation between human agency and technological constraint in digital societies
Art · Data science · Human–computer interaction · Performing arts · Visual arts · Big Data and Business Intelligence · Computer Science · Engineering · Ethics and Social Impacts of AI · Explainable Artificial Intelligence (XAI · Mechanical Engineering
Algorithms of Oppression
AI can be sexist and racist — it’s time to make it fair
AI models collapse when trained on recursively generated data
Algorithmic bias
Bias in computer systems
Social Data
Automation bias
Capturing the Complexity in Advanced Technology Use
Toward a Theory of Interactive Media Effects (Time)
Dissecting racial bias in an algorithm used to manage the health of populations
Ethics and discrimination in artificial intelligence-enabled recruitment practices
An Algorithmic Assessment of Parole Decisions
When the machine learns from users, is it helping or snooping
Algorithmic gender bias
Learning to like TikTok . . . and not
Black box measures? How to study people’s algorithm skills
To be or not to be algorithm aware
Tracing the bias loop
A sociotechnological-system approach to AI ethics
Institutionalised distrust and human oversight of artificial intelligence
Communicating and combating algorithmic bias
Algorithmic media use and algorithm literacy
Mobile technology appropriation in a distant mirror
Knowledge of automated journalism moderates evaluations of algorithmically generated news
Reinvention mediates impacts of skin tone bias in algorithms
How do people react to AI failure? Automation bias, algorithmic aversion, and perceived controllability
Machine heuristic
Rise of Machine Agency
Reinvention in the Innovation Process
TikTok and the algorithmic transformation of social media publics
What do we know about algorithmic literacy? The status quo and a research agenda for a growing field
Automating Inequality
The importance of algorithm skills for informed Internet use
Psychological factors underlying attitudes toward AI tools
Algorithmic bias
Experiencing Algorithms
| Unique citing works | 1 |
|---|---|
| Citations per year | 1 |
| Citation span | 2026 - 2026 (1) |
| Citation velocity | current |
| Highly cited | No |