KFF Health Misinformation Tracking Poll: Artificial Intelligence and Health Information

Key Findings

• The latest KFF Health Misinformation Tracking Poll finds that about two-thirds of adults say they have used or interacted with artificial intelligence (AI), though a smaller share – about one-third – say they do so at least a few times a week. Most adults (56%) are not confident that they can tell the difference between what is true and what is false when it comes to information from AI chatbots. Even among those who use or interact with AI, half say they are not confident in their ability to tell fact from fiction when it comes to information from chatbots.
• About one in six adults (17%) say they use AI chatbots at least once a month to find health information and advice, rising to one quarter of adults under age 30 (25%).
• Most adults – including a majority (56%) of those who use or interact with AI – are not confident that health information provided by AI chatbots is accurate. While about half of the public say they trust AI chatbots, such as ChatGPT, Microsoft CoPilot, or Google Gemini, to provide reliable information on practical tasks like cooking and home maintenance and on technology, fewer say they trust chatbots to provide reliable health (29%) or political information (19%). Even among those who use AI, small shares say they trust chatbots to provide reliable information about health (36%) or politics (24%).
• For most of the public, the verdict is still out on whether AI is doing more to help or doing more to hurt people trying to find accurate health information online. About one in five adults (23%) say AI is doing more to hurt those seeking accurate health information while a similar share (21%) say it is doing more to help those efforts. However, a majority of the public (55%) – including half of those who use or interact with AI (49%) – say they are unsure of the impact of AI on health information seekers.

A Third of the Public, and Nearly Half of Younger Adults, Say They Use or Interact with AI At Least Several Times A Week

Amidst the growing use of artificial intelligence by both companies and individuals, most adults (64%) say they have used or interacted with AI, though just one in three (34%) say they do so at least several times a week, including one in ten (11%) who say they interact with or use AI several times a day. About one-third of the public (35%) say they never use or interact with AI. However, as AI features become increasingly integrated into internet search engines such as Google and social media platforms such as Facebook and Instagram, these shares may not capture adults who are unknowingly using or interacting with artificial intelligence on these platforms. Perhaps unsurprisingly, younger adults ages 18-29 (47%) and those with a college degree (39%) are more likely than their counterparts to say they use or interact with AI at least several times a week.

Most Adults are Not Confident They Can Tell Whether Information from AI Chatbots Is True or False

Most U.S. adults are not confident that they can tell what is true versus what is false when it comes to information from AI chatbots, such as Chat-GPT and Microsoft Copilot. Fewer than half say they are either “very confident” (9%) or “somewhat confident” (33%) that they can tell the difference between true and false information from an AI chatbot, while a majority say they are either “not too confident” (35%) or “not at all confident” (22%). While adults who say they have used or interacted with AI are more likely than non-users to say they are at least somewhat confident in their ability to tell fact from fiction in information from AI chatbots (49% vs. 32%), even among users of this technology, half say they are not confident they can tell what is true from what is false.

Corresponding to their greater propensity to use AI platforms, younger adults are more likely than older adults to express confidence in their own ability to tell truth from fiction on these platforms. Notably, seven in ten adults ages 65 and over say they are not confident they can tell whether information is true or false on AI chatbots. This difference in confidence across age groups persists among AI users as well, with younger adults ages 18-49 who use or interact with AI reporting higher levels of self-confidence in their abilities to detect true and false information from AI chatbots than their counterparts ages 50 and older (54% vs. 41%).

One in Six Adults – and a Quarter of Young Adults – Say They Use AI Chatbots At Least Once a Month for Health Information and Advice

About one in six adults (17%) report that they use AI chatbots at least once a month to find health advice and information. Reflecting their greater use of AI overall, younger adults are more likely than older adults to say they use chatbots for health information, with a quarter (25%) of adults ages 18-29 saying they do this at least once a month.

Most Adults – Including Most AI Users – Are Not Confident That Health Information Provided by AI Chatbots is Accurate

When it comes to health information, the public is not yet convinced that AI chatbots can provide accurate information. Just one in three adults say they are “very confident” (5%) or “somewhat confident (31%) that the health information and advice they may come across on AI chatbot platforms is accurate. About six in ten adults – including a majority (56%) of AI users – say they are “not too confident” or “not at all confident” in the accuracy of health information provided by AI chatbots. Adults under age 50 and Black and Hispanic adults are somewhat more likely than those over age 50 and White adults, respectively, to say they have confidence in the accuracy of health information from AI chatbots, though about half or more across age and racial and ethnic groups say they are not confident.

Though the public is skeptical of health information from AI chatbots, larger shares say they trust AI bots to provide reliable information about other topics, including technology and practical tasks like cooking or home maintenance. Just over half of adults say they have at least “a fair amount” of trust in AI chatbots to provide reliable information about practical tasks (54%), while nearly half trust them to provide reliable information about technology (48%). However, about three in ten adults trust AI bots to provide reliable health information (29%) and one in five trust them for information about politics (19%). Trust is higher among those who use AI, with at least six in ten AI users saying they trust chatbots to provide reliable information about practical tasks and technology. Still, small shares of AI users say they trust chatbots to provide reliable information about health (36%) or politics (24%).

The Public Is Uncertain Whether AI Will Help or Hurt People Trying to Find Accurate Health Information

At this early stage in the development of consumer-facing, generative AI models, many are uncertain if these technologies are having a positive or negative impact on those seeking health information online. Over half (55%) of adults say they are not sure if AI is doing more to help or to hurt those who are trying to find accurate health information and advice online, while about one in five say they think it is doing more to help (21%) and a similar share say it is doing more to hurt (23%) these efforts. Younger adults under age 30 are divided on whether AI is doing more to help (31%) or hurt (30%) those trying to find health information and advice online, with 38% saying they are unsure; whereas majorities of adults over age 30 say they are unsure of the impact of AI when it comes to people trying to find accurate health information.

Notably, even among the people who are utilizing this technology, there is still a lot of skepticism, as three in ten AI users (30%) think it is doing more to help people trying to find accurate health information while one in five (21%) say it is doing more to hurt those efforts and about half of AI users (49%) are unsure.