SYDE Research Seminar - Values in the Responsible Application of AI in Occupational Healthcare: A Multi-Stakeholder Analysis and Comparison with the Ethics Guidelines for Trustworthy AI
Date: Thursday, September 17, 2026
Time: 9:00 AM EDT
Location: Online via Microsoft Teams
Presenter
Max Tijhuis is a PhD candidate at HumanTotalCare, in the Department of Research and Development, and at Amsterdam UMC, in the Department of Public and Occupational Health. His research focuses on the responsible use of AI in occupational healthcare, with a specific focus on ethical, communicative, and legal aspects.
Abstract
AI has received increasing attention in healthcare, including occupational healthcare, where it offers significant opportunities. With the development of AI, ethical questions quickly emerged, prompting the development of several regulatory frameworks, such as the EU AI Act, which build on the Ethics Guidelines for Trustworthy AI. While efforts have been made to translate the principles of trustworthy AI into guidelines for general healthcare, translating these principes to occupational healthcare can be challenging due to differences in stakeholder interests.
Goals:
To explore the values of employees, employers and occupational healthcare professionals that play a role in responsible application of AI in occupational healthcare. Secondly, to compare stakeholder values with the seven key requirements of the Ethics Guidelines for Trustworthy AI. Thirdly, to examine how stakeholder perspectives implicitly reflect on these key requirements within the context of occupational healthcare.
Methods:
Semi-structured interviews were conducted with Dutch employees (n=8), employers (n=7), and occupational healthcare professionals (n=8). Using values coding, we identified stakeholder values and compared them with the seven key requirements through a deductive coding framework. Findings 13 different categories of stakeholder values were found. Occupational health professionals placed the strongest emphasis on professional autonomy, non‑maleficence, professionalism/quality of care, and beneficence. Employers mostly emphasised employee autonomy, privacy/safety, transparency and role responsibility. Employees prioritised privacy, autonomy and trust, with intelligibility as a crucial component. Our data closely aligned with two of the seven key requirements (technical robustness and safety, transparency), while the other five partially align, and several values are not mentioned in the guidelines, i.e. recognition, independence and authenticity.
Conclusion:
Insight in values of employees, employers and occupational healthcare professionals can support the responsible use of AI in occupational healthcare by highlighting differences in stakeholder values and revealing that the Ethics Guidelines for Trustworthy AI lack essential contextual information for this domain.