.png){kind=icon}
.png){kind=icon}
Return
Making problems: interdisciplinary collaboration and AI ethics
.png){kind=icon}
OA
AI
DOI：10.1080/09505431.2025.2584759.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The impacts of AI and algorithmic systems on society are often spoken of as a grand societal challenge and it is increasingly understood that cultivating more fair, accountable, or transparent uses of AI, will require the inclusion of more diverse voices (social scientists, philosophers, users) in the development process. Yet, within the emerging field of AI Ethics, it is frequently observed that these disciplines speak different languages. What computer scientists mean by ‘fair’ (expressed mathematically) does not necessarily correspond to legal, philosophical or colloquial definitions of the word. Such disagreements are often presented as a barrier to problem solving. Yet others argue that disciplinary differences, including divergences in the meaning of words, are rather an opportunity: the frictions created when ways of knowing rub up against each other are necessary to develop new problem formulations. However, literature from STS around care in interdisciplinarity, suggests that these frictions must be handled carefully – attending to relations between parties. Through collective reflections on an interdisciplinary project combining the social and computer sciences, this paper illustrates how the framing of the problem evolvedas a consequenceof the curation of disciplinary disjunctures. The paper contributes to STS studies of interdisciplinarity in practice by thinking through how to recognize and foster productive disciplinary disjunctures with care. This has implications for how the AI Ethics community and policy makers can think about addressing grand societal challenges through interdisciplinary research.
Keywords:
Interdisciplinarity
AI ethics
data visualization
collaboration
problems
care
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
S
IF:
2.4
Papers:
65
Citations:
1.2K