Health plans and providers cannot control where the conversation starts but they can help determine what happens next
October 2026 | By Lindsey Miller
3 Takeaways for Health Plan Leaders:
- Members are using consumer AI to research health questions and interpret coverage.
- Trust and accuracy remain challenges as AI-generated answers influence decisions without always reflecting a member’s clinical or coverage context.
- Connected operations help close the gap between AI-generated information and accurate answers through coordinated workflows and human oversight.
At the ACHP Symposium in Washington, DC, one of the most consequential AI discussions was not about how health plans are using the technology. It was about how members and patients are already using it themselves.
People are turning to tools such as ChatGPT and Gemini to research symptoms, understand diagnoses, prepare questions, compare options, and make sense of complex health information. In many cases, they share details they might hesitate to raise with another person. The interface is immediate, conversational, and available at the exact moment a question or concern arises.
Health plans and providers cannot control whether a member begins a healthcare journey with an AI tool. They can decide whether they are prepared to help that member evaluate the information, connect it to an appropriate source, and determine what should happen next.
The Trust Paradox Is Already Visible
People do not need to feel entirely comfortable with AI to rely on it. That tension is already visible in consumer research. A 2025 survey from the Annenberg Public Policy Center found that 63 percent of people who search online for health information considered AI-generated health information somewhat or very reliable. At the same time, 49 percent of adults said they were uncomfortable with providers using AI tools rather than their experience alone when making decisions about care.
A 2025 NORC survey found that 27 percent of adults had already used AI to obtain health information at least once. Respondents identified source transparency, professional validation, the ability to compare information, and privacy assurances as factors that could increase their trust.
The pattern matters more than whether consumers describe themselves as enthusiastic about AI. People may remain skeptical while still using the technology, sharing personal information, and allowing its responses to influence the questions they ask or the actions they consider. Health plans and providers need to plan around that behavior rather than wait for attitudes to become more settled.
The Member Journey Will Not Wait
Consumer AI is changing where the healthcare journey can begin. A member may arrive at a provider appointment with a proposed diagnosis, a list of treatment questions, or a summary created from personal health details. A member may contact a health plan after asking an AI tool to interpret a denial, explain a benefit, or compare coverage options. The generated answer may be useful, incomplete, outdated, or wrong, but it has already shaped the conversation.
This creates a practical responsibility for healthcare organizations. A dismissive response may send the person back to the same tool without better context. An uncritical response may reinforce inaccurate information. The more useful role is to help the member validate the answer, understand its limits, and connect it to the clinical, benefit, or coverage information that applies to the individual.
This is not an argument that every health plan or provider must build a consumer chatbot. It is an acknowledgment that AI-mediated health decisions are becoming part of the environment in which care and coverage are delivered.
Expectations for Speed and Accuracy Are Changing
At the symposium, I heard another implication raised: AI will compress the amount of time consumers are willing to wait for an answer. A member who can receive an immediate explanation from a general-purpose AI tool may have less patience for delayed or inconsistent information about a referral, prior authorization, benefit, or bill.
The AI response may lack the member’s full context, but it still establishes an expectation of immediacy. Health plans and providers will need to respond by improving the operational layer behind the experience. Call centers, member communications, clinical teams, and digital channels need access to current information and workflows that can move a question toward resolution without repeatedly asking the member to start over.
Integration Does Not Mean Uncontrolled Access
Accepting that members will use AI does not mean connecting protected health information directly to public tools. Health plans and providers still have obligations related to privacy, security, consent, data minimization, and appropriate use. Those requirements should shape any connection between consumer AI and the healthcare system.
A practical response begins with authoritative information. Health plans can make benefit, network, coverage, payment, and service information easier for members to find and understand. Providers can make education and follow-up materials clearer and more accessible. Both can identify the source and date of information so members can distinguish current guidance from an answer generated without the right context.
Secure, member-authorized data access may also become more important. When data is shared, the member should understand what is being accessed, why it is needed, what the technology may do with it, and how to withdraw permission. Sensitive questions and decisions should have a clear path to a qualified person. The objective is to create a safer connection between the place where the question begins and the people and systems equipped to resolve it.
Health Plans Can Shape What Happens Next
Health plans have an opportunity to prepare their operations for questions and decisions that begin outside their own channels. Service teams may need to recognize when a member is working from AI-generated information. Digital experiences may need to provide clearer evidence, sources, and next steps. Workflows may need to carry the member’s question and relevant context across service, clinical, enrollment, payment, or appeal processes.
This is where connected operations and orchestration become important. The value is not in adding an AI response to another isolated portal. It is in helping the member move from general information to an accurate, accountable resolution while maintaining appropriate oversight. For Softheon, this reinforces the importance of connected member experiences, secure data exchange, and workflows that can carry context from a question through resolution.
The consumer models and access points will continue to change. The underlying behavior is unlikely to disappear. Members have found a new way to ask sensitive questions and gather information. Healthcare organizations do not need to control every conversation, but they should be prepared to provide trusted information, secure connections, and human judgment when the conversation reaches them.
If your team is exploring how to connect AI, member services, and core operations, let’s talk about what comes next.
