TL;DR
Get health and wellness essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A report on research into older adults’ technology use says AI chatbots and voice assistants can make health information easier to access, but their answers can be wrong or misleading. The researcher recommends “informed skepticism”: checking sources and quality features, protecting personal information and not treating chatbot output as a substitute for professional care.
A report on older adults’ use of artificial intelligence for health information says chatbots and voice assistants can help explain medical topics, but users should check their answers rather than trust them automatically. The researcher calls this approach “informed skepticism”, pointing to risks that include inaccurate guidance, privacy exposure and users mistaking a confident-sounding response for reliable advice.
The report, published by Medical Xpress, draws on research into how people interact with technology, including older adults’ use of voice assistants and AI chatbots. It describes tools such as ChatGPT, Claude, Gemini and health-provider-specific systems as quick ways to ask questions, get a plain-language explanation or summarize complex information. The report says those features can be appealing when a health appointment is weeks or months away or when a person finds a sensitive topic difficult to raise with another person.
In a 2025 survey of 300 adults cited by the report, about one in five respondents said they used AI chatbots for health questions. A separate poll conducted by the researcher and colleagues found that half of older adults reported using AI chatbots or voice assistants, and about 14% of that group accessed health information through them. These figures describe the surveys cited in the report; the source does not establish that they represent all adults or all older adults.
The report also cites a finding that four in five users in the 300-person survey said they never verified a health chatbot’s output. That leaves room for errors to go unnoticed. The researcher says chatbots can misclassify conditions, provide incorrect information during emergencies or respond too agreeably—an issue AI specialists call “sycophancy.” The report recommends using available source and information-quality features, checking claims against reliable published information, and avoiding personally identifying details in prompts.
The Risks Behind Convenient Answers
Fast, accessible answers could help people make sense of health information, particularly when they have ongoing conditions or new diagnoses and want explanations at home. The report says this can also give older adults more control over how they seek information. But an answer delivered instantly is not necessarily accurate, and polished language can make a faulty response harder to recognize. Convenience and credibility are separate questions.
The survey finding that many users do not check chatbot output highlights a practical concern: people may rely on a response without knowing its basis or limitations. A mistaken answer could lead someone to misunderstand symptoms or delay seeking care. The report does not measure how often chatbot errors cause harm, so it does not establish the scale of that risk. Its recommendation is to treat AI as a source to evaluate, not as a replacement for a clinician.
Privacy adds another consideration. The report says public chatbot providers may retain prompts to train future models and relays researchers’ advice not to include personally identifiable information. People searching for sensitive health topics may also want to know whether an AI feature is active in their search engine and whether it can be turned off.
Products Worth Considering
Bowflex PR1000 Home Gym
Levoit Eleva 300X HEPA Air Purifier for Home and Bedroom
Coway Airmega 250 Air Purifier for Large Rooms
As an affiliate, we earn on qualifying purchases.
How Older Adults Use AI
The report challenges the assumption that older adults are uniformly unfamiliar with technology. It says older people may be less likely than younger people to use AI overall, while also describing use of voice assistants such as Siri and Alexa for health questions. The researcher’s work suggests some older users prefer to start and finish a search on the same device. Younger adults in the cited research were more likely to treat AI as one step before consulting another source.
That difference may affect how readily users verify a response. If checking an answer means switching devices or formats, some people may not do it. The report says mainstream chatbots commonly display warnings that their answers can contain mistakes, but those notices may be easy to miss. It also notes that search engines may include AI-generated summaries directly in results, meaning users can encounter AI even when they did not set out to open a chatbot.
The report points to the Aging with AI Act, introduced in the U.S. Senate in 2026, as a policy development that reflects growing attention to potential benefits and risks. It does not provide further details about the bill’s status or provisions.
““informed skepticism””
— The researcher whose work is described in the Medical Xpress report
What the Surveys Cannot Establish
The survey figures are based on specific samples cited in the report, including one survey of 300 adults. The report does not give enough information here to determine how representative those samples are, or whether the findings apply across different age groups, locations and types of chatbot. The cited numbers should not be read as population-wide estimates without that context.
The report also does not quantify how often health chatbots give incorrect advice, how many users are harmed or whether checking responses reliably prevents harm. It describes concerns and user habits, not a clinical trial showing that chatbot use improves health outcomes. The Senate bill’s progress and any resulting policy changes are also not specified.
Practical Checks for Users
The report’s immediate advice is to learn how to disable AI features where possible, inspect a chatbot’s source or quality information, and check health claims against reliable published sources. It notes that Google users can add “-ai” to a search query to remove AI content from results, while some services use different opt-in or opt-out settings. Features may vary, so users should check the controls of the service they are using.
For health concerns, chatbot output should not be treated as a diagnosis or as a substitute for care from a qualified health professional. The report calls for more research into how older adults use AI and the benefits and risks involved. The next policy milestone is not clear from the source; the Aging with AI Act was introduced in 2026, but no later action is reported.
Key Questions
What does “informed skepticism” mean when using a health chatbot?
It means using a chatbot as a possible source of explanation while checking the quality and credibility of its response instead of assuming it is correct.
How many older adults in the cited poll used AI?
The researcher’s 2025 poll with colleagues found that one in two older adults reported using AI chatbots or voice assistants. About 14% of that group accessed health information. The report does not establish that these figures represent all older adults.
Can an AI chatbot replace a doctor?
The report describes chatbots as tools for accessing and explaining information, while warning that they can give inaccurate guidance or misclassify conditions. Their output should not be treated as a diagnosis or a replacement for care from a qualified health professional.
What privacy steps does the report recommend?
Researchers cited in the report generally advise users not to enter personally identifying information into chatbot prompts. The report says public chatbot providers may retain prompts for training future models.
How can Google users remove AI content from a search?
The report says adding “-ai” to the end of a Google search query will remove AI content from results. Search features and controls can change, so users should check the settings available in their service.
Source: medicalxpress.com
