How Generative AI in Health Care Is Changing Patient Expectations

Key Takeaways

• Patients benchmark doctors against AI chatbots for empathy
• AI‑drafted messages reduce clinician documentation load
• Unintegrated AI advice creates a shadow record outside EHR
• Governance and workflow integration are critical to maintain trust

Pulse Analysis

The rise of generative AI in health care reflects a broader consumer‑tech wave where patients expect instant, plain‑language explanations. Studies such as the 2023 JAMA Internal Medicine analysis reveal that many patients perceive chatbot replies as more empathetic than physician communication, a perception that reshapes the doctor‑patient power dynamic. Health systems are responding by piloting AI‑assisted drafting of portal messages, aiming to cut documentation time while delivering consistent, patient‑focused language. However, these tools operate without full access to the medical record, raising concerns about accuracy and the potential for conflicting advice.

Beyond efficiency, the unchecked proliferation of AI‑generated content creates a parallel "shadow record"—patient‑driven narratives and recommendations that remain invisible to clinicians. This fragmented data stream can undermine clinical decision‑making, as physicians may miss early warning signs or spend additional time reconciling discrepancies. Legal liability also becomes murkier; while clinicians retain professional accountability, AI tools lack regulatory responsibility, leaving institutions to navigate new risk landscapes. The tension between perceived empathy and actual clinical authority forces providers to confront how trust is earned and maintained in an AI‑augmented environment.

Strategic integration offers a path forward. Robust governance frameworks must differentiate AI fluency from clinical responsibility, ensuring that AI suggestions are clearly flagged for clinician review. Workflow designs should enable seamless capture of patient‑generated AI interactions into the electronic health record, turning the shadow record into actionable data. Investment in training, audit trails, and transparent communication can preserve the therapeutic relationship while leveraging AI to reduce friction and surface early health signals. Ultimately, health systems that proactively embed AI with clear guardrails will likely enhance patient satisfaction and safety, whereas passive adoption risks widening mistrust and operational fragmentation.