One in four Americans now turns to AI chatbots like ChatGPT for health information, and while most use it to prepare for doctor visits, a significant minority are skipping appointments altogether based on AI-generated advice. A nationally representative survey of over 5,500 U.S. adults conducted between October 2025 and December 2025 by the West Health-Gallup Center on Healthcare in America found that 25% of Americans have used an AI tool or chatbot for health information or advice, with roughly 14 million adults reporting they did not see a healthcare provider because of AI-generated health information they received .

How Are Americans Actually Using AI for Health Information?

The picture that emerges from the data is nuanced. Most Americans who use AI for health purposes treat it as a supplement to traditional care rather than a replacement. Among those who have recently used AI for health information, 59% say they use it to research on their own before visiting a doctor, and 56% use it to research after a doctor visit . This suggests that for the majority, AI serves as a preparation or reflection tool that potentially makes healthcare interactions more informed and focused.

The types of health questions Americans ask AI reveal practical, everyday concerns. Over half of recent users report asking AI about nutrition or exercise questions (59%), and a similar share ask about physical symptoms (58%). Beyond these basics, 46% have used AI to understand medication side effects, 44% to interpret medical information, and 38% to research a diagnosis or medical condition . These patterns suggest people are using AI to bridge knowledge gaps and prepare for conversations with providers.

Why Are Some Americans Skipping Doctor Visits Entirely?

While supplemental use dominates, the minority who skip provider visits based on AI advice represents a meaningful public health concern. The 14% of recent AI users who report skipping a provider visit translates to approximately 14 million U.S. adults when projected to the entire adult population . This raises questions about whether AI is appropriately substituting for medical care in certain situations.

The reasons people turn to AI instead of seeing providers vary significantly by income and access. Among adults in households earning less than $24,000 annually, 32% say they have used AI because they could not pay for a doctor's visit, compared with just 2% among those earning $180,000 or more . Similarly, 16% of recent AI users cite inability to access a provider, and 21% report using AI because they felt dismissed or ignored by a provider in the past. These findings suggest that for some Americans, AI fills gaps created by cost, geography, or poor prior experiences with healthcare systems.

Steps to Use AI Responsibly for Health Information

• Cross-reference with professional guidance: Use AI to research and prepare questions before seeing a provider, rather than making final healthcare decisions based solely on AI responses.
• Assess trust levels honestly: Only 4% of AI users strongly trust the accuracy of AI-generated health information; consider this low confidence level when making decisions about your care.
• Watch for safety red flags: About 11% of AI users report receiving health advice they believed was unsafe; if AI recommends something that seems risky or contradicts your provider's guidance, seek professional clarification.
• Use AI for symptom exploration, not diagnosis: AI can help you understand what questions to ask your doctor, but it should not replace a professional diagnosis or treatment plan.

The trust gap in AI-generated health information is striking. Among those who have used AI for health information in the past 30 days, roughly one-third say they trust it (33%), one-third neither trust nor distrust it (33%), and one-third distrust it (34%). However, only 4% say they strongly trust the accuracy of AI-generated health information, suggesting that many Americans are making healthcare decisions based on information they lack confidence in . Additionally, about 11% of recent users report that AI recommended healthcare information or advice they believed was unsafe.

Age also shapes how Americans use AI for health research. Younger adults are significantly more likely than older adults to use AI for self-directed research before seeing a doctor. Among recent users aged 18 to 29, 69% say they use AI to research on their own before seeing a doctor, compared with 43% of those aged 65 and older . Despite this age gap, self-directed research remains prevalent among older adults, with more than four in 10 aged 65 and older using AI for this purpose.

On the positive side, nearly half of Americans who have used AI for healthcare information (46%) say the AI tool or chatbot made them feel more confident when talking with or asking questions of a provider . Additionally, 22% report that AI helped them identify issues earlier, and 19% say it helped them avoid unnecessary medical tests or procedures. These benefits suggest that when used appropriately, AI can enhance patient engagement and potentially improve healthcare outcomes.

The most frequently reported AI tools for health information are general conversational systems such as ChatGPT or Copilot (61%), followed by AI tools embedded within web searches, such as Google AI summaries (55%) . This indicates that people are relying on widely available, consumer-facing AI systems rather than specialized medical AI applications.

As AI becomes more woven into how Americans navigate healthcare, the challenge for the healthcare system is clear. For some patients, AI is already serving a productive role, helping them clarify questions and review medical information in a system facing time constraints and workforce pressures. However, the 14 million Americans who skip provider visits based on AI advice, combined with the low trust levels in AI accuracy, suggests that healthcare systems will need to adapt to how these tools are being incorporated into the healthcare journey. Whether AI tools can appropriately substitute for certain healthcare interactions, and under what circumstances, remains an important question as use of these tools continues to grow .