Shadow AI in Swedish Health Care: Qualitative Analysis of Physicians' Free-Text Answers

Clinical Snapshot

Key Findings

• P Value: Not applicable — qualitative study

• Effect Size: Not applicable — qualitative study; no quantitative effect sizes reported

• Primary Outcome: Four thematic categories of Shadow AI use identified: (1) clinical work and decision-making (differential diagnoses, rare case support, second opinion); (2) administrative work (patient communication, clinical documentation); (3) research and professional development (staying current, knowledge access); (4) technological interest and curiosity (exploring generative AI capabilities). An overarching professional paradox was identified: Shadow AI simultaneously poses regulatory and safety risks while addressing real gaps in institutional clinical and administrative support.

• Nnt Or Sensitivity: Not applicable — qualitative content analysis; no diagnostic accuracy, therapeutic effect, or prognostic metrics reported. The study's analytic yield is thematic rather than statistical.

• Confidence Interval: Not applicable — qualitative study

Clinical Application

The study does not evaluate an intervention; it describes existing behaviour. Its clinical application lies in informing governance rather than clinical practice change. Health system leaders, clinical informaticists, and medical educators can use these findings to design sanctioned AI pathways that address the identified gaps — particularly in clinical decision support for rare or complex cases and in administrative burden reduction — thereby reducing the drivers of Shadow AI adoption without relying solely on prohibition. The findings are highly relevant to Australian health care. The Australian Digital Health Agency's National Digital Health Strategy and the Therapeutic Goods Administration's (TGA) Software as a Medical Device (SaMD) regulatory framework create a comparable regulatory tension to the EU MDR context described in this study. General-purpose AI tools (e.g., ChatGPT) are not TGA-listed as medical devices, rendering their clinical use analogous to the 'Shadow AI' described here. The RACGP and specialist colleges have not yet issued definitive guidance on clinician use of unsanctioned AI tools. Australian hospitals and primary health networks face the same paradox: physicians are likely using these tools to fill gaps in clinical decision support and documentation, yet no sanctioned pathways exist for most use cases. State health departments and the Australian Commission on Safety and Quality in Health Care should consider this evidence when developing AI governance frameworks. The PBS and MBS do not currently fund AI-assisted clinical decision support, creating an additional structural driver for Shadow AI adoption. Physicians working in publicly funded, digitally integrated health systems who have access to internet-connected devices during clinical work. Findings are most directly applicable to hospital-based and primary care physicians in high-income countries with established electronic health record infrastructure. The four thematic categories (clinical decision support, documentation, professional development, curiosity) are likely to resonate across specialties and seniority levels.

Abstract

BACKGROUND: The rapid emergence of artificial intelligence (AI) has outpaced its formal adoption in health care organizations, contributing to the emergence of Shadow AI, defined here as the use of unauthorized AI tools by medical professionals. Under the European Union Medical Device Regulation, AI tools used for clinical purposes must undergo conformity assessment before use; general-purpose tools such as ChatGPT have not done so, rendering their clinical application unauthorized at the regulatory level. While Shadow AI offers potential efficiency gains and higher performance, it poses significant risks to data privacy, clinical safety, and regulatory compliance. Despite its growing prevalence, empirical research on the purposes for which physicians use Shadow AI remains scarce. OBJECTIVE: This study explores the purposes for which physicians describe using Shadow AI in their work. METHODS: We conducted a cross-sectional survey of physicians employed in Swedish health care organizations (N=357; response rate~64%). Data were collected between December 2023 and January 2024 via a verified online panel. We conducted a qualitative content analysis of free-text responses on the use of unauthorized AI tools. We applied theoretical lenses from the sociology of professions and paradox theory to interpret the empirical findings. RESULTS: Physicians use Shadow AI for several purposes, which we grouped into 4 categories: clinical work and decision-making, administrative work, research and professional development, and technological interest and curiosity. More specifically, Shadow AI is used as a colleague and second opinion for clinical decision support (eg, differential diagnoses and rare cases), administrative tasks such as patient communication and documentation, and research aimed at staying up to date and exploring developments in generative AI. Physicians described using these tools compensated for perceived gaps in institutional systems, reducing workload, and accessing knowledge considered difficult to obtain through conventional channels. The findings reveal a tension between physicians' drive to improve their practice and the regulatory and organizational constraints that render such use unauthorized. CONCLUSIONS: Shadow AI used by physicians presents both opportunities and risks for health care professionals and organizations. Shadow AI indicates gaps where formal hospital systems may fail to meet health care professionals' needs and signals a way for physicians to strengthen their experience-based knowledge. It represents a renegotiation of professional boundaries, as physicians bypass institutional constraints to maintain professional efficacy. The findings highlight a paradox in which the same tools that pose regulatory and safety risks also address real gaps in clinical and administrative support, suggesting that governance approaches must account for this tension rather than relying on prohibition alone.

References

1. 1.Petersson, L., Irgang, L., Mauritzon, I., & Holmén, M. (2026). Shadow AI in Swedish health care: Qualitative analysis of physicians' free-text answers. Journal of Medical Internet Research. https://doi.org/10.2196/93484