ACM Journal on Responsible Computing | Volume 2, Number 4
Abstract
Digital technologies are becoming intimately interwoven with people’s daily lives [Van de Poel et al. 2022; Van Est et al. 2014]. As these technologies are transforming and reconfiguring cognitive, affective, bodily and conative capacities, the field of computing shows a growing interest in how this entanglement of human life with digital technology requires analyses that foreground human vulnerability. For instance, Greenberg and Marble [2023] argue for the essential importance of accounting for vulnerability in what they refer to as “Person-Machine Teaming”. Relatedly, Tschopp [2020] argues in favor of putting vulnerability at center stage in the context of trustworthy AI and human-machine interaction. Furthermore, Van Riemsdijk [2018] proposes that “Intimate Computing” is best understood as “Computing with Vulnerability”, aiming at realizing Intimate Technologies that empower and support people in engaging with and reflecting on their (technology-mediated) personal vulnerabilities throughout their day. We see additional pleas for a vulnerability-focused understanding of our digitally-mediated lives in Malgieri [2023] and Ranchordas
Study specs
The study reviews and synthesizes existing literature and frameworks addressing vulnerability in human-technology interactions, including concepts like 'Intimate Computing' and 'Person-Machine Teaming'.
- Institution
- University of Texas,Microsoft Research,Google DeepMind,Google,University of Washington,World Economic Forum
- Discipline
- Ethics,Governance of Computing Research,AI Ethics
- Study Type
- Literature Review
- Year
- 2025
- Human Data Platform
- Prolific
- Source
- View Source Google Scholar
Measured Outcomes
Human vulnerability in the context of digitally-mediated interactions and the role of computing frameworks in addressing them.
Peer Review & Critical Discussion
Potential Selection Bias in 2023 Cohort
The participant pool shows a concerning overrepresentation of users from high-income demographics. Looking at Table 3, we can see that 78% of respondents had annual incomes above $75k, which significantly limits the generalizability of these findings to broader populations.
Non-naive Participants Issue
I've noticed a methodological concern regarding participant naivety. Given that Prolific users often complete multiple studies, there's a real risk that participants had prior exposure to similar experimental paradigms, which could confound the results.
RLHF Applicability to This Study Design
The implications for RLHF training pipelines are understated. If we accept the authors' conclusions about preference stability, this has direct consequences for how we should structure reward model training. The temporal decay effect described in Section 4.2 is particularly relevant.
Verify your expertise to join discussion
Create an account and verify your credentials to participate in peer discussions.