Patients Are Asking Chatbots First—And It Signals a System in Trouble
A new commentary in Annals of Internal Medicine argues that the surge in chatbot use for health questions is not simply a story about risky technology. Instead, it reflects structural failures in how patients access care—turning instant, always-on AI systems into the default front door for medical information. The piece highlights recent survey findings: 32% of U.S. adults used AI for health advice or information in the past year.
Robert B. Shpiner, MD, a clinical professor of pulmonary and critical care at UCLA, frames the behavior as a rational response to friction. Patients report that they cannot get timely appointments, lack a consistent primary care clinician, face affordability barriers, and doubt whether the health system will respond effectively. In that vacuum, chatbots become a rapid substitute for both guidance and reassurance.
The commentary does not deny safety concerns. It acknowledges that AI outputs can be inaccurate, overconfident, or poorly contextualized, creating potential harm when used as a medical decision aid. However, Shpiner cautions that “risk warnings” alone are insufficient—because warnings don’t address why people are driven to consult AI in the first place.
Technically, chatbot adoption functions like a real-time triage interface: users enter symptoms or concerns and receive immediate conversational responses. But the commentary suggests the underlying problem is that patients experience slow feedback loops elsewhere, including billing complexity, prior authorization delays, and difficult portal communication.
In this view, AI’s popularity is a diagnostic signal of an overloaded and fragmented care pathway. When patients struggle to navigate coverage rules or obtain clear answers through traditional channels, they rationally seek alternatives that provide continuity and responsiveness.
Shpiner argues that policy and governance around AI—while important—won’t meaningfully reduce chatbot-related risks unless health systems also fix the access and communication bottlenecks that funnel people toward AI assistance. In other words, the safest chatbot strategy may require building a better healthcare “front end.”
The commentary emphasizes governance as downstream: ethical guardrails and monitoring cannot compensate for upstream barriers that leave patients with few workable options. Until access is streamlined and responsiveness improves, chatbot usage may continue to grow regardless of cautionary messaging.
The headline takeaway is therefore double-edged: AI can be both a symptom of healthcare strain and a contributor to misinformation risk. The solution, Shpiner suggests, is not only technical oversight but systemic redesign of how patients get help.
Web references and full article details are embargoed until release.
Subject of Research: People
Article Title: Why Patients Ask the Chatbot First
News Publication Date: 28-Jul-2026
Web References: http://dx.doi.org/10.7326/ANNALS-26-01819
References: —
Image Credits: —
Keywords
Artificial intelligence; Generative AI; Public health
Tags: AI health advice reliancebarriers to timely healthcare accesschatbot use as primary medical information sourceconsequences of healthcare affordability barriersimpact of primary care shortagespatient behavior in healthcare decision-makingpotential harms of AI-driven health guidancerisks of inaccurate AI health outputsrole of AI in medical triagesafety concerns of medical chatbotsstructural healthcare system failuressystemic causes of healthcare technology dependence
