Dr. Sachin Jain, a physician and healthcare executive, recalls a moment from his medical training that still resonates: patients rarely asked the questions they’d prepared overnight, because by the time the clinical team arrived, the moment had passed.
From Notebooks to AI Assistants
Decades later, patients are solving that problem themselves. Armed with smartphones and AI assistants like ChatGPT and Claude, they’re recording visits and later reconstructing what was said, asking follow-up questions, and tracking medications and appointments.
Yet the gap between clinician-controlled tools and patient-facing ones remains wide. OpenNotes gave patients access to clinical notes, but a recent patient still asked to record the visit. Access alone didn’t satisfy the need for agency—the ability to question, compare, and continuously engage with one’s own care.
But the technology raises new concerns. Generative AI tools are “often wrong but never in doubt,” making it hard for patients, and doctors, to distinguish accurate answers from misleading ones.
Read Also: Startup raises $22M to expand at-home HPV screening
Still, the deeper issue isn’t legal risk. It’s accountability. The medical chart is the version of the visit doctors control. A recording is the version the patient controls. That shift unsettles many clinicians, not just because of liability, but because it demands ownership of what was actually said.
The Weight of Ownership
The tension remains unresolved. Patients continue recording, questioning, and prompting their devices long after appointments end. Clinicians, meanwhile, grapple with a new form of accountability, one that demands ownership of spoken words, not just signed notes.
