
Written by: Mayo Clinic Staff
AI can answer health questions in seconds. But should you trust it with your symptoms? Here's what to know before you rely on it.
Imagine you've been feeling tired for weeks. Your usual strategies, like rest and extra coffee, aren't helping. Before deciding whether to schedule an appointment with a healthcare professional, you open an artificial intelligence (AI)-powered chat tool and type, "What health conditions cause fatigue?"
Within seconds, a list of answers appears. It includes stress, anemia, thyroid issues, depression, chronic illness and cancer. The information feels organized and sounds accurate — and a little scary.
But is this answer trustworthy? And what should you do with this information?
AI-created health information is widely used
Nearly 8 in 10 adults in the U.S. turn to the internet for answers to health questions. Instead of scrolling through websites, many people find answers in the AI-generated summary that appears at the top of their search results. (1)
But it doesn't stop there. More than 1 in 5 adults worldwide are turning directly to AI chatbots, like ChatGPT and Gemini, to ask health questions. (2) It's fast, convenient and free. But Mayo Clinic experts warn that AI-generated information isn't always reliable or accurate.
Why you can't always trust AI
When it comes to using AI for health information, there are a few key limitations to keep in mind:
1. Diagnosing and treating illness is too complex for a machine
AI tools don't have access to your full medical record — and you shouldn’t upload or share it with them. These tools can't examine you or run tests the way a healthcare professional can. They don't have the ability to reason like a human or explain how they came to a conclusion. (3) These qualities are necessary for making safe and accurate medical decisions.
2. AI can be wrong, even when it sounds confident
AI chatbots give answers based on patterns in data. They don't "know" facts in the way a health professional does. Sometimes AI information sounds true but is completely incorrect. This is known as hallucination. (4, 5) For example, when asked how to get more minerals from food, AI has been known to recommend eating rocks. (6)
3. AI-created information may be biased
AI systems are trained on large amounts of data that may contain bias or gaps. That means it may not reflect everyone's experience fairly. (7, 8) For example, an AI system that learned from information about people in the United States and parts of Europe might miss signs of depression. That's because it doesn't know that in some cultures, people show sadness through physical symptoms like headaches or tiredness, rather than talking about their feelings.
4. AI spreads misinformation
AI doesn't know what's true and what's not. (6) It may pull answers from flawed or misleading sources it finds online. When people see alarming health stories online, they often forward or repost them — even if they know the information may not be true. As false information spreads, there’s more of it online. AI may then repeat that false information. (9)
5. People and AI don't communicate well together
One study found that people seeking health information don't tend to give AI chatbots enough specific information for clear, accurate answers. (4) And small differences in how symptoms are described can completely change the answer from AI, making it less accurate. (4,5) For example, when two people asked about the same symptoms but used different words, AI told only one of them the correct answer, which was to get emergency care. (4)
How to use AI more safely and effectively
If you still want to use AI to learn about a health topic, here are practical steps to reduce risk:
- Use AI for general education, not diagnosis. AI is best suited for explaining medical terms or giving general wellness advice. For example, you might ask, "What does hypertension mean in plain language?" or "How can I add more movement into my day?"
- Cross-check everything. Verify information with trusted sources, like websites for Mayo Clinic or the American Medical Association. Most importantly, review what you learn with your healthcare team. (6)
- Ask clear, specific questions. Instead of asking, "Is coughing a bad sign?" try, "What are common causes of chronic dry cough in adults?" Clear, focused questions tend to produce more useful and balanced answers. (6) Remember: Healthcare professionals are trained to ask the right follow-up questions. AI isn't. (4)
- Protect your privacy. Don't share personal information, like your full name, date of birth, address, medical records or insurance details. Even health details should be shared cautiously, especially on public or free platforms. (6)
The bottom line
Think of AI as a research assistant, not as your healthcare professional. It can be a helpful tool for summarizing ideas or getting big-picture information. AI can be extremely useful in preparing questions to ask your care team. But when it comes to your health, the safest and most effective decisions are still made with a trusted healthcare professional.
Related links:
References
- Many in U.S. consider AI-generated health information useful and reliable. Annenberg Public Policy Center. https://www.annenbergpublicpolicycenter.org/many-in-u-s-consider-ai-generated-health-information-useful-and-reliable/. Accessed Feb. 10, 2026.
- Yun HS, et al. Online health information-seeking in the era of large language models: Cross-sectional web-based survey study. Journal of Medical Internet Research. 2025; doi:10.2196/68560.
- Ullah E, et al. Challenges and barriers of using large language models (LLM) such as ChatGPT for diagnostic medicine with a focus on digital pathology — A recent scoping review. Diagnostic Pathology. 2024; doi:10.1186/s13000-024-01464-7.
- Bean AM, et al. Reliability of LLMs as medical assistants for the general public: A randomized preregistered study. Nature Medicine. 2026; doi:10.1038/s41591-025-04074-y.
- Giorgi S, et al. Evaluating generative AI responses to real-world drug-related questions. Psychiatry Research. 2024; doi:10.1016/j.psychres.2024.116058.
- What doctors wish patients knew about using AI for health tips. American Medical Association. ama-assn.org/practice-management/digital-health/what-doctors-wish-patients-knew-about-using-ai-health-tips. Accessed Feb. 17, 2026.
- Yoon SC, et al. Digital psychiatry with chatbot: Recent advances and limitations. Clinical Psychopharmacology and Neuroscience. 2025; doi:10.9758/cpn.25.1346.
- Thakkar A, et al. Artificial intelligence in positive mental health: A narrative review. Frontiers in Digital Health. 2024; doi:10.3389/fdgth.2024.1280235.
- Saeidnia HR, et al. Generative AI and health misinformation: Production, propagation, and mitigation — A systematic review. BMC Public Health. 2026; doi:10.1186/s12889-025-26148-9.
Connect
@heyjoe415 --- I used AI for medical information. It/he/her was just as complete as the dr. I had seen. In fact maybe a little more. AI had time to explain questions, and was correct in "its" replies. Even had a sense of humor and thanked me. But I sure would never ever give out a S.S. number or personal data on line. I don't bank on line either.
-
Like -
Helpful -
Hug
1 Reaction@mermaid7272 I only bank and shop on our desktop which has the bells and whistles security. Never on my MS Surface nor my phone. As far as AI, I have read AI explanations but always seek a second and sometimes third source to verify.
-
Like -
Helpful -
Hug
1 ReactionMermaid and Linda,
On the topic of online banking (and AI), please always use two-factor authentication to log in. This usually requires a text sent to your phone after you have entered your password. AI can now figure out even difficult passwords with ease, so a second-layer of security/verification is advised.
And Mermaid I agree with you on AI for medical information. I don't rely on it but it is useful a lot of the time. I do verify the information with my Dr.
My best to both of you.
Joe
-
Like -
Helpful -
Hug
2 Reactions@heyjoe415 the two factor verification is used for our banking account on our desktop. I do not trust phone or tablet security - so I never access such information on those devices.
-
Like -
Helpful -
Hug
2 Reactions@walkinggirl
Confounded😤. Banks, insurance companies, government, religious instiutions, department stores, on line sales all want us to use some form of computer/social media interface. They are closing down mailrooms.
Same organizations then notify me "We are so sorry. Hackers broke our saran wrap firewall and got all your private medical and financial information. Soon your accounts will be drained. Sorry"
Even the government can't stop 'em.
-
Like -
Helpful -
Hug
2 ReactionsThanks Linda.
In the age of AI, it's probably best to do what you're doing, access sensitive information from only one device. And in all cases, use two-factor authentication.
Authentication keys are also increasing in popularity. Apple has one and it only allows access to an account from a fingerprint scan. It's a good idea to use this biometric account access wherever it is available. But like all things, verify its safety first.
Joe
-
Like -
Helpful -
Hug
2 Reactions@heyjoe415 While all of this is great security bells and whistles, older people (we are talking 80-year-olds here), are increasingly frustrated by all of this. It's no wonder that we/they become increasingly alienated and easily targeted by fraud and scams. Trust - believing what we are seeing (taking the time to check the URL carefully, spelling, proper grammar, going to the actual URL on one's own) has become increasingly more time consuming. I think that goes along with depending on Dr AI and Dr Google as well as some of the things we hear on the news based on nonscientific ideas. Here on Connect, we are very concerned about false information and advice given without knowing particular histories, especially when it impacts health and well-being. Verifying health advice and information with known trustworthy sites (Mayo, Cleveland Clinic etc) should be an automatic response.
-
Like -
Helpful -
Hug
4 Reactions@walkinggirl Good points.
An issue I have with biometrics is that there's no Plan B. If your fingerprint gets hacked, what can you do?
You can change a password but you can't change a fingerprint.
-
Like -
Helpful -
Hug
2 Reactions@colleenyoung
There is another problem with AI, that is specific to organizations like Mayo:
AI "engines" scour the Internet for information to build their massive databases of knowledge. Unfortunately, often these engines are not able to tell the difference between some desperate person posting about some unproven treatment, & real medical science.
Google web search in particular is guilty of this. It will often post some statement as truth, where the source is a group discussion like we have here, or a YouTube video which at best biased or worse.
Not much you can do about the latter, but the former is a problem in medical discussions. If several people start to endorse an unproven treatment in a thread, & an AI engine uses that thread as the source for reporting the value of such a treatment, it can look like the unproven treatment (taken out of context) is endorsed by the medical organization hosting the discussion.
This is a fine line for medical group discussions hosted by medical organizations like Mayo. If the organization blocks such discussions, people needing help may go elsewhere, rather than learning the truth about unproven treatments. On the other hand, permitting unproven treatments to go without sound science rebuttal, can create the appearance of support, & that impression can be quickly propagated worldwide.
Personally, I lean toward removing such discussions, as I believe that the danger is greater in letting such discussions persist without prompt rebuttal.
However, I am not in charge.
-
Like -
Helpful -
Hug
4 Reactions@readandlearn, you raise a very important point that people should be aware of. AI engines scrape everything, including comments in online forums like Reddit, Facebook, and health forums like this one and others such as Inspire, HealthUnlocked, PatientsLikeMe and countless others.
When reading an AI overview in search results, it is important to know that the information can be wrong. AI tools can sound sure, even when they’re wrong. Always check the original sources. See more guidance provided in this blog post:
- What is Generative AI? What does this mean on Mayo Clinic Connect? https://connect.mayoclinic.org/blog/about-connect/newsfeed-post/what-is-generative-ai-artificial-intelligence-what-does-this-mean-on-mayo-clinic-connect/
We take moderation of Mayo Clinic Connect seriously, removing disinformation, correcting misinformation, and helping people understand the dis- and misinformation they may find elsewhere. We also moderate posts (and sometimes remove them) if they contravene the Community Guidelines (https://connect.mayoclinic.org/blog/about-connect/tab/community-guidelines/)
You can read more about why and how we moderate Mayo Clinic Conncet here: https://connect.mayoclinic.org/blog/about-connect/tab/moderators/
While moderators monitor discussions, we also rely on our members to tell us about any inappropriate content or behavior. Like you have done (thank you).
If you see a post that you think is inappropriate, please report it immediately. f you see a post that requires the attention of the moderating team, please let us know. Report it.
How to report a post:
1. Click the three dots (…) in the lower right corner of the comment.
2. Select Report discussion or comment.
3. Write the reason for the report and click SUBMIT.
-
Like -
Helpful -
Hug
10 Reactions