Systematic Review of Generative AI in Healthcare: Automation Bias and Deskilling Concerns

Generative AI improves diagnostic accuracy and workflow efficiency but risks automation bias (reported in 10/29 studies) and clinician deskilling (9/29), urging balanced integration with human oversight and explainable AI.

• Of 29 included studies, 10 reported automation bias, 9 deskilling, and 9 impacts on diagnostic reasoning.
• Mitigation strategies include human-in-the-loop frameworks, explainable AI (XAI), ethical governance, and continuous reskilling.
• Recommendations emphasize redefining clinician-AI interaction as a cognitive partnership rather than passive reliance.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.