Earning Trust in Health AI: Governance, Cybersecurity, and Workforce Readiness for Ethical and Safe Adoption

Document Type

Conference Proceeding

Publication Date

6-2026

Abstract

Research Objective: Artificial intelligence (AI) is rapidly entering healthcare, yet trust, cybersecurity readiness, and ethical governance remain critical determinants of safe and equitable adoption. This study examined how professionals working at the intersection of cybersecurity, data science, engineering, and healthcare perceive the opportunities and risks of AI in health settings, with attention to trustworthiness, system reliability, patient protection, and the conditions required for responsible implementation.

Study Design: This qualitative study used semi-structured interviews to explore real-world experiences, expectations, and concerns regarding AI integration in healthcare. Interviews were transcribed and analyzed using reflexive thematic analysis to identify patterns related to trust, cybersecurity practices, ethical considerations, and implications for clinical care, workforce functioning, and patient well-being.

Population Studied: Participants included cybersecurity professionals, AI and data engineers, healthcare technology experts, and individuals supporting clinical systems or health data infrastructures (n = 20). Most participants held advanced degrees and lived in urban or suburban environments. The sample was predominantly Black/African American, with additional racial and ethnic diversity represented. All participants had direct exposure to digital health systems, cybersecurity implementation, or AI-related decision-making.

Principal Findings: Participants acknowledged meaningful potential benefits of AI, including enhanced efficiency, improved clinical decision support, automation of routine tasks, earlier detection capabilities, and improved patient monitoring and safety. However, trust in AI was consistently described as fragile and conditional. Four overarching themes emerged: (1) trust is earned, not assumed, with transparency, accountability, and demonstrated reliability shaping willingness to rely on AI; (2) cybersecurity and governance gaps undermine confidence, including concerns about data breaches, unclear standards, uneven regulation, and inadequate readiness of health environments to host AI technologies; (3) human oversight remains essential, with strong concern about replacing rather than supporting clinicians and emphasis on safeguards that protect patients and reinforce ethical responsibility; and (4) workforce and system readiness shape impact, as limited training, policy uncertainty, variable understanding of AI risks, and weak infrastructure threaten safe integration. Participants stressed that AI can improve healthcare but only when implemented intentionally, transparently, and with patient-centered safeguards.

Conclusions: Although AI holds significant potential to transform healthcare, trust cannot be presumed. Trustworthy AI depends on strong cybersecurity, ethical governance, transparency, human oversight, and systems designed to protect patients while supporting clinicians. Without clear standards and investment in readiness, AI risks amplifying vulnerability rather than advancing care.

Implications for Policy or Practice: Findings identify actionable priorities to guide responsible AI adoption in healthcare: strengthening AI governance frameworks; implementing robust cybersecurity protections and continuous monitoring; promoting transparency to build patient and provider confidence; investing in workforce training and ethical preparedness; ensuring meaningful human oversight; and aligning AI deployment with patient safety, fairness, and accountability. These insights provide practical guidance for policymakers, healthcare leaders, technologists, and clinicians seeking to advance trustworthy, equitable AI that improves care while safeguarding patients and communities.

Comments

AcademyHealth Annual Research Meeting, May 30-June 2, 2026, Seattle, WA

Link to Full Text

Share

COinS