The Regulatory Blind Spot Putting European Patients at Risk: Why Primary Care AI Is Escaping Medical Device Oversight

A quiet transformation is underway in European primary care clinics, one that operates largely beneath the radar of regulatory oversight. General-purpose artificial intelligence systems—tools originally designed for consumer applications—are increasingly being deployed to summarize patient records, suggest diagnoses, and support clinical decision-making. These implementations are occurring without the regulatory clearances, clinical validation, or safety mechanisms required of medical devices under European law. The result is a dangerous regulatory blind spot that exposes patients to unquantified risks while creating unfair competitive dynamics in the healthcare AI marketplace.

Healthcare AI expert Ángel Alberich-Bayarri has sounded the alarm on this emerging crisis, arguing that existing EU Medical Device Regulation (MDR) frameworks already cover AI-driven clinical software—and that regulatory shortcuts represent both a patient safety threat and a fundamental breach of fair competition principles. The implications extend far beyond regulatory technicalities, touching on questions of clinical liability, market integrity, and the future governance of artificial intelligence in healthcare settings.

The Scope of Unregulated Clinical AI Deployment

The deployment of general-purpose AI tools in European primary care settings has accelerated rapidly, particularly in Spain and other southern European countries. These systems, built on large language model architectures similar to ChatGPT, are being positioned as "assistive tools" rather than medical devices—a distinction that allows them to bypass the rigorous regulatory pathways required for software that influences clinical decision-making.

The clinical applications are far from trivial. Healthcare providers are using these tools to synthesize complex patient histories, identify diagnostic possibilities from symptom presentations, and generate treatment recommendations. In primary care settings—where clinical uncertainty is highest and diagnostic decisions have cascading consequences throughout the healthcare system—these AI interventions directly influence patient outcomes.