Research shows that one in seven Americans trust AI chatbots as much as they trust a doctor. A survey involving 1,000 millennial parents and an additional 1,000 individuals from the general population looked into how people utilize AI for health management, whether for themselves or their children, and their understanding of how AI generates its recommendations.
Interestingly, among the respondents who use AI for health-related concerns, nearly five times as many trust AI at the same level as they do their doctors, compared to those who don’t use AI. Additionally, 60% indicated they would accept AI-generated health advice even if it went against what a healthcare provider said.
Conducted by Talker Research for MD Live by Evernorth, the survey revealed that almost two-thirds (63%) of AI users have followed health advice from AI without consulting a healthcare provider first. This level of confidence likely stems from the way AI communicates; half of the participants stated that if AI provided a clear or reassuring answer regarding their health, they would likely trust it without a second thought.
Those employing AI for health matters typically pose around nine health-related queries each week. They often choose to rely on AI because it’s conveniently accessible at any hour (37%), offers quicker responses than a doctor’s visit (31%), and provides a sense of reassurance regarding health worries (30%). Yet, about 28% of users admit they verify AI answers less than half the time.
For millennial parents using AI in health scenarios, 58% acknowledged taking action based on AI-generated advice for their children without prior confirmation from healthcare professionals.
Moreover, many respondents expressed comfort in depending solely on AI for various health decisions related to their offspring, including assessing minor symptoms (80%), determining needed care (71%), and obtaining medication advice (71%).
Despite this trust in AI, many users overlook how their input can affect the accuracy of AI responses. About 78% didn’t realize that the wording of their questions can influence the answers they receive. In fact, more than half (53%) admitted to rephrasing their health questions when the initial AI responses didn’t meet their expectations, with 27% saying they do this all the time. Many users also tend to frame questions in a way that leads AI toward answers they want to hear.
Interestingly, 38% describe their symptoms to see what might be wrong. About a third inquire if their symptoms could relate to an existing condition they suspect (34%) or seek confirmation of a suspected diagnosis (32%).
“This survey indicates that people are often consulting AI in hopes of a specific answer—whether that’s reassurance or validation of a worry,” noted Dr. Maggie Williams, medical director of virtual primary care at MD Live by Evernorth. “This approach is quite different from how a physician typically engages a patient, where detailed questions and context gathering play crucial roles before arriving at a conclusion.”
Moreover, many Americans seem to have an inflated perception of how AI operates. One in five individuals believes that AI exclusively draws from trusted medical sources (22%), while the same percentage of regular users wrongly assume that AI incorporates their personal medical history, even when that information has not been shared.
Simultaneously, countless users recognize that they don’t provide a comprehensive overview of their health when seeking advice. For instance, 42% have even fabricated details about who they were asking for health information due to privacy concerns.
Over three-quarters of habitual users (76%) leave out their medical backgrounds or personal data for similar reasons, with 81% attributing it to privacy issues.
This lack of full disclosure can lead to users receiving responses that aren’t entirely relevant to their situations. Nearly 60% (59%) later realized that the health information they got from AI didn’t actually apply to them, and almost half (49%) acted on that information before realizing it wasn’t suitable.
Even with the increasing use of AI for health inquiries, respondents expressed worries regarding receiving incorrect advice (34%) and the absence of human oversight (31%).
Looking to the future, one in four indicated they plan to incorporate AI into more health decisions by 2027, while 27% stated they would trust AI-generated guidance more if a doctor reviewed it.
Dr. Williams highlighted that these results show a clear trend of individuals turning to AI for health-related questions, which is a positive development as it encourages greater engagement in personal health. However, she emphasized that healthcare providers play a vital role by bringing clinical insights and understanding to the conversation, ensuring that all relevant details are considered before a conclusion is made. It’s important to explore how AI and healthcare professionals can collaborate to present the clearest and most comprehensive information to patients while maintaining necessary safeguards and oversight.
