By Dennis Thompson HealthDay ReporterFRIDAY, Oct. 9, 2026 (HealthDay News) \u2014 Want to buy an accurate blood pressure monitor?<\/p>\n
Don\u2019t turn to AI for help, a new study warns.<\/p>\n
Three of four leading AI-powered search tools frequently recommended home BP monitors that haven\u2019t been validated for accuracy, according to findings presented Thursday at the American Heart Association\u2019s annual meeting on hypertension.<\/p>\n
\u201cWe found that most AI tools performed only slightly better than if you had flipped a coin for each question,\u201d said researcher Dr. Anna Soriano<\/a>, a third-year internal medicine resident at the University of Montreal in Canada.<\/p>\n Buying the wrong blood pressure monitor could have consequences for your health, Soriano noted in a news release.<\/p>\n \u201cUsing that device may result in inaccurate blood pressure readings, which could lead to an inappropriate diagnosis or treatment decisions,\u201d Soriano said.<\/p>\n High blood pressure<\/a> affects nearly 47% of adults in the U.S., more than 125 million people. Only about 1 in 4 of those have their blood pressure controlled to a healthy range, according to the American Heart Association (AHA).<\/p>\n Many people are prescribed a home BP monitor to help their healthcare team track their high blood pressure, and might turn to AI to find a good device, researchers said.<\/p>\n For the new study, researchers asked Google Gemini, ChatGPT, Copilot and Perplexity about 324 different home blood pressure monitors, of which 179 had not been validated as accurate.<\/p>\n The team asked the AI whether each device had been validated, and even asked specific questions directing the AI to check named validation registries.<\/p>\n Results showed that Copilot, ChatGPT and Perplexity delivered accurate recommendations only 63% to 83% of the time. Google Gemini had the best results, getting it right 86% to 91% of the time.<\/p>\n \u201cEven Google Gemini, which performed best, was often wrong and couldn\u2019t find information that is easily located,\u201d Soriano said. \u201cPeople may unknowingly think a device is validated based on the AI tool\u2019s inaccurate responses.\u201d<\/p>\n Researchers could not say why the AI tools struggled to identify devices that had been validated and were in online registries as such.<\/p>\n \u201cIt is surprising, and almost counterintuitive, that AI tools had so much difficulty specifically identifying validated devices, since those devices are the ones with clear listings on official registries,\u201d Soriano said. \u201cIn many cases, the AI tools found the blood pressure monitor was listed within an official website, however, it seems AI was not able to interpret being listed on the registry as proof of validation.\u201d<\/p>\n The study revealed potential shortcomings of AI that are a caution people using it to choose medical devices, AHA spokesman Dr. Keith Ferdinand<\/a> said in a news release.<\/p>\n \u201cIn addition, home blood pressure devices need to be both validated \u2014 listed on the national registry validatebp.org \u2014 and accurate,\u201d continued Ferdinand, who is chair of preventive cardiology at Tulane University in New Orleans. \u201cWith the proper technique and regular monitoring, readings from home BP devices are a valuable component of integrated, individual treatment plans that can improve patient care and outcomes.\u201d<\/p>\n The AHA meeting continues through Sunday in Arlington, Virginia. Findings presented at medical meetings should be considered preliminary until published in a peer-reviewed journal.<\/p>\n SOURCE: American Heart Association, news release, Oct. 8, 2026<\/p>\n