Public comfort with AI in health care is slipping, and that could slow adoption
An Ohio State survey reported by EurekAlert suggests public comfort with AI in health care has fallen. The finding matters because even technically strong tools can stall if patients and families do not trust the systems behind them.
Patient trust is becoming one of the key constraints on health AI. A declining comfort level could slow deployment in exactly the places where AI advocates want rapid scaling: triage, diagnostics, scheduling, and clinical documentation.
That dynamic should not surprise anyone. Patients are being asked to accept systems that can be opaque, error-prone, and difficult to question, often in settings where they already feel vulnerable. Without clear explanations and visible safeguards, skepticism is rational.
The survey result also matters for hospitals because adoption is no longer just a vendor decision. If patients perceive AI as a cost-cutting substitution rather than a quality-enhancing aid, institutions may face reputational risk alongside technical risk.
The lesson for the field is straightforward: trust cannot be bolted on after rollout. If health systems want durable adoption, they need to communicate where AI is used, what humans still do, and how errors are detected and corrected.