Stefan Buijsman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Stefan Buijsman has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Bart Knijnenburg
School of Computing, Clemson University
2 shared publications - Daricia Wilkinson
Microsoft Research
2 shared publications - Helma Torkamaan
Delft University of Technology, Delft, Netherlands
2 shared publications Juan-Pablo Bermúdez
University of Southampton, Southampton, UK
2 shared publications- Luciano Cavalcante Siebert
Delft University of Technology The Netherlands
2 shared publications - Mohammad Tahaei
UC Berkeley’s International Computer Science Institute
2 shared publications
Latest Publications
- Open Access
- 24 April 2026
- Published by Springer Nature in Philosophy & Technology
- Vol. 39 (2) , 84
What is Wrong With Automation Bias?
- 24 April 2026
- Published by Springer Nature in Philosophy & Technology
- Vol. 39 (2) , 84
Open AccessWhat is Wrong With Automation Bias?
- Open Access
- 27 February 2026
- Published by Springer Nature in Minds and Machines
- Vol. 36 (1) , 14
Accuracy is not all you need! The Reasons to Require AI Explainability
Do we need explanations of AI outputs in order to use AI systems (in high-risk settings)? This question has been actively debated recently, with one group denying that explanations are needed as long as the AI system is sufficiently accurate. What matters, according to them, is that outcomes improve. The other group argues that we have procedural reasons, centered around autonomy and self-advocacy, which trump outcome-based arguments to the contrary. I here present a set of arguments to show that outcome-based arguments should in fact also favor explainability for many of the current systems, as challenges with human oversight and accountability often lead to worse overall outcomes even if a more accurate AI system is integrated. Critics of explainability overlooked the fact that AI operates within a broader socio-technical system, and its accuracy alone tells us little of the final outcomes. In addition, I consolidate the procedural arguments and present a view of the upshot of these arguments. On this, we should avoid applications of AI that largely replace decision-making (relegating humans to the position of checking outputs). We can, however, use AI in other roles even for high-risk decision making while conforming to all of the requirements set by both outcome-based and procedural arguments. What matters, in the end, is the ability to explain decisions, and with the right role for AI that is possible even when supported by opaque systems.
- 27 February 2026
- Published by Springer Nature in Minds and Machines
- Vol. 36 (1) , 14
Open AccessAccuracy is not all you need! The Reasons to Require AI Explainability
Do we need explanations of AI outputs in order to use AI systems (in high-risk settings)? This question has been actively debated recently, with one group denying that explanations are needed as long as the AI system is sufficiently accurate. What matters, according to them, is that outcomes improve. The other group argues that we have procedural reasons, centered around autonomy and self-advocacy, which trump outcome-based arguments to the contrary. I here present a set of arguments to show that outcome-based arguments should in fact also favor explainability for many of the current systems, as challenges with human oversight and accountability often lead to worse overall outcomes even if a more accurate AI system is integrated. Critics of explainability overlooked the fact that AI operates within a broader socio-technical system, and its accuracy alone tells us little of the final outcomes. In addition, I consolidate the procedural arguments and present a view of the upshot of these arguments. On this, we should avoid applications of AI that largely replace decision-making (relegating humans to the position of checking outputs). We can, however, use AI in other roles even for high-risk decision making while conforming to all of the requirements set by both outcome-based and procedural arguments. What matters, in the end, is the ability to explain decisions, and with the right role for AI that is possible even when supported by opaque systems.
- 3 January 2026
- Published by Springer Nature in Philosophy & Technology
- Vol. 39 (1) , 8
Inauthentic Value Shifts: More than Manipulation
- 3 January 2026
- Published by Springer Nature in Philosophy & Technology
- Vol. 39 (1) , 8
Inauthentic Value Shifts: More than Manipulation
- Open Access
- 13 November 2025
- Published by BMJ in BMJ Digital Health & AI
- Vol. 1 (1) , e000086
Responsible artificial intelligence in healthcare: a systematic review on the use of ethical principles in the development and deployment of artificial intelligence
Objective: As hospitals increasingly adopt artificial intelligence (AI) to manage rising patient volumes, workforce shortages and healthcare costs, concerns about ethical implementation have become prominent. This systematic review aims to assess how hospital-focused AI literature addresses the WHO’s six ethical AI principles—autonomy; well-being and safety; transparency and explainability; responsibility and accountability; inclusiveness and equity; and responsiveness and sustainability. Methods and analysis: A systematic review (PROSPERO registration: CRD42022347871) was conducted by searching Embase, MEDLINE ALL, Web of Science and the Cochrane Central Register of Controlled Trials from inception to December 2023, supplemented by Google Scholar. English-language studies describing AI (machine learning, deep learning, predictive analytics) relevant to inpatient settings and referencing at least one WHO principle were included. Two reviewers independently screened titles, abstracts and full texts, extracting data on publication year, country, study design, AI type, technology readiness level and ethical considerations. Discrepancies were resolved by consensus. Results: Of 4770 unique records, 673 were included. Most (83%) originated from high-income countries, with publication volume rising sharply after 2021. Of these, 558 (83%) addressed at least one WHO principle in depth, most frequently inclusiveness and equity (49%), transparency and explainability (45%) and autonomy (42%). Well-being and safety (26%) and responsibility and accountability (29%) were less frequently covered, while responsiveness and sustainability (6%) was rarely explored. Among 44 studies developing AI applications with technology readiness levels 1–6, ethical principles were acknowledged but rarely operationalised. Conclusion: Hospital-based AI research demonstrates increasing attention to ethical principles but lacks comprehensive application, particularly regarding sustainability. High-income countries dominate this discourse, underscoring the need for broader global engagement. To achieve equitable, safe and sustainable AI in clinical practice, clearer operational guidance and more inclusive collaboration is warranted.
- 13 November 2025
- Published by BMJ in BMJ Digital Health & AI
- Vol. 1 (1) , e000086
Open AccessResponsible artificial intelligence in healthcare: a systematic review on the use of ethical principles in the development and deployment of artificial intelligence
Objective: As hospitals increasingly adopt artificial intelligence (AI) to manage rising patient volumes, workforce shortages and healthcare costs, concerns about ethical implementation have become prominent. This systematic review aims to assess how hospital-focused AI literature addresses the WHO’s six ethical AI principles—autonomy; well-being and safety; transparency and explainability; responsibility and accountability; inclusiveness and equity; and responsiveness and sustainability. Methods and analysis: A systematic review (PROSPERO registration: CRD42022347871) was conducted by searching Embase, MEDLINE ALL, Web of Science and the Cochrane Central Register of Controlled Trials from inception to December 2023, supplemented by Google Scholar. English-language studies describing AI (machine learning, deep learning, predictive analytics) relevant to inpatient settings and referencing at least one WHO principle were included. Two reviewers independently screened titles, abstracts and full texts, extracting data on publication year, country, study design, AI type, technology readiness level and ethical considerations. Discrepancies were resolved by consensus. Results: Of 4770 unique records, 673 were included. Most (83%) originated from high-income countries, with publication volume rising sharply after 2021. Of these, 558 (83%) addressed at least one WHO principle in depth, most frequently inclusiveness and equity (49%), transparency and explainability (45%) and autonomy (42%). Well-being and safety (26%) and responsibility and accountability (29%) were less frequently covered, while responsiveness and sustainability (6%) was rarely explored. Among 44 studies developing AI applications with technology readiness levels 1–6, ethical principles were acknowledged but rarely operationalised. Conclusion: Hospital-based AI research demonstrates increasing attention to ethical principles but lacks comprehensive application, particularly regarding sustainability. High-income countries dominate this discourse, underscoring the need for broader global engagement. To achieve equitable, safe and sustainable AI in clinical practice, clearer operational guidance and more inclusive collaboration is warranted.
- 1 November 2025
- Published by Springer Nature
- p. 1-16
Conceptual Disconnect: The Prospect of Concept-Based AI Explanations
- 1 November 2025
- Published by Springer Nature
- p. 1-16
Conceptual Disconnect: The Prospect of Concept-Based AI Explanations
- Open Access
- 18 October 2025
- Published by Springer Nature in Philosophy & Technology
- Vol. 38 (4) , 1-20
Is Meaningful Human Control Over Personalised AI Assistants Possible? Ethical Design Requirements for The New Generation of Artificially Intelligent Agents
Recently, several large tech companies have pushed the notion of AI assistants into the public debate. These envisioned agents are intended to far outshine current systems, as they are intended to be able to manage our affairs as if they are personal assistants. In turn, this ought to give users a leg up, as one prominent tech exec has put it. However, it remains to be seen how these Personal AI Assistants (PAIAs) are implemented, and critical reflection on how and whether they can be implemented in a responsible way is needed. Currently, such agents are undertheorized and this may cause us to misunderstand their value and capacity. In this paper, we explore and critique the potential for responsible implementation by considering some design requirements based on the notion of meaningful human control. If we desire to have control over such assistants, then we need to be able to do so meaningfully and effectively. In looking at the design requirements, we run into the issue that their broad and differing capacities make any kind of design requirements hard because there are simply no standards to which we can measure PAIAs. Furthermore, it seems that the implementation of these assistants will be a matter of trade-offs both in capacities and in values, which will likely lead to enhancement for some rather than an improvement for all.
- 18 October 2025
- Published by Springer Nature in Philosophy & Technology
- Vol. 38 (4) , 1-20
Open AccessIs Meaningful Human Control Over Personalised AI Assistants Possible? Ethical Design Requirements for The New Generation of Artificially Intelligent Agents
Recently, several large tech companies have pushed the notion of AI assistants into the public debate. These envisioned agents are intended to far outshine current systems, as they are intended to be able to manage our affairs as if they are personal assistants. In turn, this ought to give users a leg up, as one prominent tech exec has put it. However, it remains to be seen how these Personal AI Assistants (PAIAs) are implemented, and critical reflection on how and whether they can be implemented in a responsible way is needed. Currently, such agents are undertheorized and this may cause us to misunderstand their value and capacity. In this paper, we explore and critique the potential for responsible implementation by considering some design requirements based on the notion of meaningful human control. If we desire to have control over such assistants, then we need to be able to do so meaningfully and effectively. In looking at the design requirements, we run into the issue that their broad and differing capacities make any kind of design requirements hard because there are simply no standards to which we can measure PAIAs. Furthermore, it seems that the implementation of these assistants will be a matter of trade-offs both in capacities and in values, which will likely lead to enhancement for some rather than an improvement for all.
- Open Access
- 6 October 2025
- Published by Springer Nature in Philosophy & Technology
- Vol. 38 (4) , 1-4
The Role of Digital Literacy in Maintaining Autonomy in AI Decision-Support: Balancing the Burdens
Integrating AI systems into workflows risks undermining the competence of the people supported by them, specifically due to a loss of meta-cognitive competence. We discuss a recent suggestion to mitigate this through better uncertainty quantification. While this is certainly a step in the right direction, there is a question whether users are sufficiently supported to engage in critical reflection with literacy and tools alone. We therefore suggest that socio-technical system design focused on the role of AI systems is crucial to preserving autonomy, even when supported by uncertainty quantification.
- 6 October 2025
- Published by Springer Nature in Philosophy & Technology
- Vol. 38 (4) , 1-4
Open AccessThe Role of Digital Literacy in Maintaining Autonomy in AI Decision-Support: Balancing the Burdens
Integrating AI systems into workflows risks undermining the competence of the people supported by them, specifically due to a loss of meta-cognitive competence. We discuss a recent suggestion to mitigate this through better uncertainty quantification. While this is certainly a step in the right direction, there is a question whether users are sufficiently supported to engage in critical reflection with literacy and tools alone. We therefore suggest that socio-technical system design focused on the role of AI systems is crucial to preserving autonomy, even when supported by uncertainty quantification.
- 24 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Integrating Participatory Research and Literature Review to Identify Social and Ethical Requirements for Responsible AI in Heart Failure Treatment
The rapid advancement of Artificial Intelligence (AI) in healthcare raises significant social and ethical concerns, particularly in cardiovascular care, the leading cause of death worldwide. Addressing these challenges requires a multidisciplinary, multi-stakeholder approach to ensure the responsible development and implementation of AI-driven solutions. First, a literature review examined how key ethical principles - namely, autonomy, confidentiality, data privacy, and equal treatment - are integrated into AI applications for heart failure care. These findings informed an online workshop with patient representatives, clinicians, and experts on ethical, legal, and social issues to discuss ethical concerns and practical implications. A team of requirements engineers and digital health experts synthesized insights from the literature review and workshop, specifying an initial set of requirements. Following multiple iterations with a multidisciplinary review team, the final set of 25 requirements was established. These were translated into software or system specifications where applicable or used to define guidelines for real-world implementation. This study provides a structured approach to embedding ethical and social considerations into AI-driven healthcare solutions. Its methods and key requirements are transferable to other AI applications in public health, promoting responsible and equitable adoption.
- 24 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Integrating Participatory Research and Literature Review to Identify Social and Ethical Requirements for Responsible AI in Heart Failure Treatment
The rapid advancement of Artificial Intelligence (AI) in healthcare raises significant social and ethical concerns, particularly in cardiovascular care, the leading cause of death worldwide. Addressing these challenges requires a multidisciplinary, multi-stakeholder approach to ensure the responsible development and implementation of AI-driven solutions. First, a literature review examined how key ethical principles - namely, autonomy, confidentiality, data privacy, and equal treatment - are integrated into AI applications for heart failure care. These findings informed an online workshop with patient representatives, clinicians, and experts on ethical, legal, and social issues to discuss ethical concerns and practical implications. A team of requirements engineers and digital health experts synthesized insights from the literature review and workshop, specifying an initial set of requirements. Following multiple iterations with a multidisciplinary review team, the final set of 25 requirements was established. These were translated into software or system specifications where applicable or used to define guidelines for real-world implementation. This study provides a structured approach to embedding ethical and social considerations into AI-driven healthcare solutions. Its methods and key requirements are transferable to other AI applications in public health, promoting responsible and equitable adoption.
- Top Cited PapersOpen Access
- 9 July 2025
- Published by Springer Nature in Philosophy & Technology
- Vol. 38 (3) , 1-23
Autonomy by Design: Preserving Human Autonomy in AI Decision-Support
AI systems increasingly support human decision-making across domains of professional, skill-based, and personal activity. While previous work has examined how AI might affect human autonomy globally, the effects of AI on domain-specific autonomy—the capacity for self-governed action within defined realms of skill or expertise—remain understudied. We analyze how AI decision-support systems affect two key components of domain-specific autonomy: skilled competence (the ability to make informed judgments within one's domain) and authentic value-formation (the capacity to form genuine domain-relevant values and preferences). By engaging with prior investigations and analyzing empirical cases across medical, financial, and educational domains, we demonstrate how the absence of reliable failure indicators and the potential for unconscious value shifts can erode domain-specific autonomy both immediately and over time. We then develop a constructive framework for autonomy-preserving AI support systems. We propose specific socio-technical design patterns—including careful role specification, implementation of defeater mechanisms, and support for reflective practice—that can help maintain domain-specific autonomy while leveraging AI capabilities. This framework provides concrete guidance for developing AI systems that enhance rather than diminish human agency within specialized domains of action.
- 9 July 2025
- Published by Springer Nature in Philosophy & Technology
- Vol. 38 (3) , 1-23
Top Cited PapersOpen AccessAutonomy by Design: Preserving Human Autonomy in AI Decision-Support
AI systems increasingly support human decision-making across domains of professional, skill-based, and personal activity. While previous work has examined how AI might affect human autonomy globally, the effects of AI on domain-specific autonomy—the capacity for self-governed action within defined realms of skill or expertise—remain understudied. We analyze how AI decision-support systems affect two key components of domain-specific autonomy: skilled competence (the ability to make informed judgments within one's domain) and authentic value-formation (the capacity to form genuine domain-relevant values and preferences). By engaging with prior investigations and analyzing empirical cases across medical, financial, and educational domains, we demonstrate how the absence of reliable failure indicators and the potential for unconscious value shifts can erode domain-specific autonomy both immediately and over time. We then develop a constructive framework for autonomy-preserving AI support systems. We propose specific socio-technical design patterns—including careful role specification, implementation of defeater mechanisms, and support for reflective practice—that can help maintain domain-specific autonomy while leveraging AI capabilities. This framework provides concrete guidance for developing AI systems that enhance rather than diminish human agency within specialized domains of action.
- 2 November 2024
- Published by Springer Nature
- p. 40-53
Defining and Evaluating the Degrees of Abstraction in Explanations with Kolmogorov Complexity
- 2 November 2024
- Published by Springer Nature
- p. 40-53
Defining and Evaluating the Degrees of Abstraction in Explanations with Kolmogorov Complexity