Thomas Ploug
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Thomas Ploug has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Søren Holm
The University of Manchester
36 shared publications Steinar Thorvaldsen
UiT The Arctic University of Norway
7 shared publicationsUlrik Sandborg-Petersen
Aalborg University
7 shared publications- Peter Øhrstrøm
Aalborg University, Aalborg, Denmark
5 shared publications Peter Øhrstrøm
4 shared publicationsSøren Holm
Centre for Social Ethics and Policy, Faculty of Law, School of Social Sciences, University of Manchester, M13 9PL Manchester, UK
4 shared publications
Latest Publications
- Open Access
- 28 October 2026
- Published by Springer Nature in BMC Medical Ethics
Best Practice Guidelines for Bioethical Use of Medical AI in Healthcare. An eDelphi Consensus Exercise
Introduction: Medical Artificial Intelligence (MAI) is transforming healthcare but raises significant ethical concerns. Despite emerging principles and regulations, to our knowledge, no standardized clinical bioethics framework exists to guide MAI practice. Objective: This Delphi study sought global expert agreement on ethical challenges, principles, guidelines, and practical tools for MAI use. Method: A four-round international eDelphi study, conducted under ACCORD and DELPHISTAR guidance, engaged multidisciplinary stakeholders identified through systematic review and targeted recruitment. Preliminary items were evidence-derived and iteratively refined. Consensus (≥ 80% agreement across two rounds) was predefined. Surveys were anonymized and randomized. Ethical approval and protocol registration were secured before study initiation. Results: The panel included 53 participants from 21 countries. Survey rounds started November 2025 and ended February 2026. 170 items were considered. 132 reached consensus. Items were classified into bioethical issues, values, principles, solutions, a checklist for clinicians, and report markers for risk management. Discussion: This study produced clinician-focused practice guidelines for the bioethical use of MAI. Ethical components aligned with prior literature, while no new bioethical principles met consensus. Limitations included geographic imbalance and stakeholder representation. Recommendations emphasized human responsibility, governance, and practical implementation tools. Conclusion: This Delphi consensus provides practical, ethics-based guidance for the responsible integration of MAI into clinical practice and institutional governance.
- 28 October 2026
- Published by Springer Nature in BMC Medical Ethics
Open AccessBest Practice Guidelines for Bioethical Use of Medical AI in Healthcare. An eDelphi Consensus Exercise
Introduction: Medical Artificial Intelligence (MAI) is transforming healthcare but raises significant ethical concerns. Despite emerging principles and regulations, to our knowledge, no standardized clinical bioethics framework exists to guide MAI practice. Objective: This Delphi study sought global expert agreement on ethical challenges, principles, guidelines, and practical tools for MAI use. Method: A four-round international eDelphi study, conducted under ACCORD and DELPHISTAR guidance, engaged multidisciplinary stakeholders identified through systematic review and targeted recruitment. Preliminary items were evidence-derived and iteratively refined. Consensus (≥ 80% agreement across two rounds) was predefined. Surveys were anonymized and randomized. Ethical approval and protocol registration were secured before study initiation. Results: The panel included 53 participants from 21 countries. Survey rounds started November 2025 and ended February 2026. 170 items were considered. 132 reached consensus. Items were classified into bioethical issues, values, principles, solutions, a checklist for clinicians, and report markers for risk management. Discussion: This study produced clinician-focused practice guidelines for the bioethical use of MAI. Ethical components aligned with prior literature, while no new bioethical principles met consensus. Limitations included geographic imbalance and stakeholder representation. Recommendations emphasized human responsibility, governance, and practical implementation tools. Conclusion: This Delphi consensus provides practical, ethics-based guidance for the responsible integration of MAI into clinical practice and institutional governance.
- Open Access
- 23 September 2026
- Published by Springer Nature in Research Square
Healthcare AI Ethics in Real-World Implementation: Implementation Findings as Ethical Evidence and Cross-Dimensional Alignment
Healthcare-AI ethics is often examined through normative principles or research already framed as ethical, overlooking ethically significant findings that emerge in real-world implementation without being explicitly framed as ethical. To address this gap, this study examined which explicit and implicit ethical concepts and considerations emerge when AI is implemented in real-world healthcare, how they are framed, where they are situated, and how they are addressed. From 298 ethically relevant extracts, seven themes and 27 subthemes were developed through a predominantly inductive thematic synthesis and subsequently interpreted through four conceptual dimensions: ethical concepts and considerations, ethical framing, ethical locus, and ethical response. Ethically significant findings concerned trust, professional agency, accountability, clinical fit and safety, equity and patient experience, perceived value, and socio-technical alignment. They were often expressed as implementation concerns rather than explicit ethical claims. Causes, consequences, responsibility, and capacity for action were frequently distributed across technical, clinical, organizational, and patient levels, while responses concentrated at the point of use, shifting compensatory work toward clinicians or patients even when problems originated upstream. The synthesis shows that empirical implementation findings can themselves constitute ethical evidence. Comparison across framing, locus, and response generated the concept of cross-dimensional alignment: whether problem framing reflects relevant conditions, responsibility corresponds to meaningful capacity for action, and responses reach conditions capable of change. Responsible healthcare AI therefore depends not only on identifying what is ethically at stake, but on whether ethical governance has practical traction on the socio-technical conditions through which benefits, burdens, risks, and responsibilities are produced.
- 23 September 2026
- Published by Springer Nature in Research Square
Open AccessHealthcare AI Ethics in Real-World Implementation: Implementation Findings as Ethical Evidence and Cross-Dimensional Alignment
Healthcare-AI ethics is often examined through normative principles or research already framed as ethical, overlooking ethically significant findings that emerge in real-world implementation without being explicitly framed as ethical. To address this gap, this study examined which explicit and implicit ethical concepts and considerations emerge when AI is implemented in real-world healthcare, how they are framed, where they are situated, and how they are addressed. From 298 ethically relevant extracts, seven themes and 27 subthemes were developed through a predominantly inductive thematic synthesis and subsequently interpreted through four conceptual dimensions: ethical concepts and considerations, ethical framing, ethical locus, and ethical response. Ethically significant findings concerned trust, professional agency, accountability, clinical fit and safety, equity and patient experience, perceived value, and socio-technical alignment. They were often expressed as implementation concerns rather than explicit ethical claims. Causes, consequences, responsibility, and capacity for action were frequently distributed across technical, clinical, organizational, and patient levels, while responses concentrated at the point of use, shifting compensatory work toward clinicians or patients even when problems originated upstream. The synthesis shows that empirical implementation findings can themselves constitute ethical evidence. Comparison across framing, locus, and response generated the concept of cross-dimensional alignment: whether problem framing reflects relevant conditions, responsibility corresponds to meaningful capacity for action, and responses reach conditions capable of change. Responsible healthcare AI therefore depends not only on identifying what is ethically at stake, but on whether ethical governance has practical traction on the socio-technical conditions through which benefits, burdens, risks, and responsibilities are produced.
- Open Access
- 17 May 2026
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 6 (1)
Operationalizing AI contestability through technical feasibility analysis
- 17 May 2026
- Published by Springer Nature in Discover Artificial Intelligence
- Vol. 6 (1)
Open AccessOperationalizing AI contestability through technical feasibility analysis
- 15 February 2026
- Published by Springer Nature
- p. 41-56
Towards Trustworthy AI: A Framework for Integrating Ethics Committees into Product Development
- 15 February 2026
- Published by Springer Nature
- p. 41-56
Towards Trustworthy AI: A Framework for Integrating Ethics Committees into Product Development
- 15 February 2026
- Published by Springer Nature
- p. 1-23
Conceptual Misalignment in XAI
- 22 December 2025
- Published by Springer Nature in AI and Ethics
- Vol. 6 (1) , 78
A review of ethical AI frameworks in product development: taking stock and moving forward
- 22 December 2025
- Published by Springer Nature in AI and Ethics
- Vol. 6 (1) , 78
A review of ethical AI frameworks in product development: taking stock and moving forward
- Open Access
- 25 June 2025
- Published by SAGE Publications in Journal of the Royal Society of Medicine
- Vol. 118 (8) , 248-252
The need for patient rights in AI-driven healthcare – risk-based regulation is not enough
- 25 June 2025
- Published by SAGE Publications in Journal of the Royal Society of Medicine
- Vol. 118 (8) , 248-252
Open AccessThe need for patient rights in AI-driven healthcare – risk-based regulation is not enough
- 24 February 2025
- Published by Taylor & Francis in The American Journal of Bioethics
- Vol. 25 (3) , 124-126
Consenting to and Contesting AI Use – Different Functions but Both Are Necessary
- 24 February 2025
- Published by Taylor & Francis in The American Journal of Bioethics
- Vol. 25 (3) , 124-126
Consenting to and Contesting AI Use – Different Functions but Both Are Necessary
- Open Access
- 15 July 2023
- Published by Springer Nature in Philosophy & Technology
- Vol. 36 (3) , 1-6
People Should Have a Right Not to Be Subjected to AI Profiling Based on Publicly Available Data! A Reply to Holm
- 15 July 2023
- Published by Springer Nature in Philosophy & Technology
- Vol. 36 (3) , 1-6
Open AccessPeople Should Have a Right Not to Be Subjected to AI Profiling Based on Publicly Available Data! A Reply to Holm
- Open Access
- 7 March 2023
- Published by Springer Nature in Philosophy & Technology
- Vol. 36 (1) , 1-22
The Right Not to Be Subjected to AI Profiling Based on Publicly Available Data—Privacy and the Exceptionalism of AI Profiling
Social media data hold considerable potential for predicting health-related conditions. Recent studies suggest that machine-learning models may accurately predict depression and other mental health-related conditions based on Instagram photos and Tweets. In this article, it is argued that individuals should have a sui generis right not to be subjected to AI profiling based on publicly available data without their explicit informed consent. The article (1) develops three basic arguments for a right to protection of personal data trading on the notions of social control and stigmatization, (2) argues that a number of features of AI profiling make individuals more exposed to social control and stigmatization than other types of data processing (the exceptionalism of AI profiling), (3) considers a series of other reasons for and against protecting individuals against AI profiling based on publicly available data, and finally (4) argues that the EU General Data Protection Regulation does not ensure that individuals have a right not to be AI profiled based on publicly available data.
- 7 March 2023
- Published by Springer Nature in Philosophy & Technology
- Vol. 36 (1) , 1-22
Open AccessThe Right Not to Be Subjected to AI Profiling Based on Publicly Available Data—Privacy and the Exceptionalism of AI Profiling
Social media data hold considerable potential for predicting health-related conditions. Recent studies suggest that machine-learning models may accurately predict depression and other mental health-related conditions based on Instagram photos and Tweets. In this article, it is argued that individuals should have a sui generis right not to be subjected to AI profiling based on publicly available data without their explicit informed consent. The article (1) develops three basic arguments for a right to protection of personal data trading on the notions of social control and stigmatization, (2) argues that a number of features of AI profiling make individuals more exposed to social control and stigmatization than other types of data processing (the exceptionalism of AI profiling), (3) considers a series of other reasons for and against protecting individuals against AI profiling based on publicly available data, and finally (4) argues that the EU General Data Protection Regulation does not ensure that individuals have a right not to be AI profiled based on publicly available data.