KFF Poll Finds Americans Are Using AI for Health Advice Faster Than Trust Is Catching Up

KFF’s latest polling on AI for health information and advice underscores a central tension in healthcare AI: adoption is rising because the tools are convenient, always available, and often easier to access than a clinician, but trust is still conditional. That combination is more consequential than simple enthusiasm or skepticism alone, because it points to a world in which people may act on AI outputs even when they are not fully confident those outputs are reliable.

For health systems, payers, and regulators, this is a warning that consumer AI is becoming part of the care pathway whether institutions explicitly endorse it or not. Patients are likely to use generative AI to interpret symptoms, lab results, insurance questions, and treatment options long before formal triage occurs. That shifts the point of influence upstream, where oversight is weakest and errors are hardest to detect.

The poll also highlights why transparency is becoming a competitive issue, not just an ethical one. Tools that clearly disclose data sources, limitations, and when they are not a substitute for professional care may ultimately earn more durable user trust than systems that optimize for fluency alone. In healthcare, confidence built on polished language without visible evidence is fragile.

The broader implication is that AI governance can no longer be framed only around clinical software purchased by hospitals. Consumer-facing AI now affects demand, expectations, and decision-making across the system. If policymakers focus too narrowly on regulated medical devices, they risk missing the layer of AI that many patients are already using as their first point of contact with health information.