AI-Powered Devices Gain Traction Among Older Adults for Independent Living, Survey Reveals

AI-Powered Smart Devices Help Older Adults Age at Home, Survey Finds

Older adults often find artificial intelligence smart home devices and voice assistants useful in helping them age in place, according to a new survey by the University of Michigan. More than half of people surveyed aged 50 and older have used generative AI tools they spoke or typed messages to, according to the National Poll on Healthy Aging, released by the U-M Institute for Healthcare Policy and Innovation. By comparison, a June Pew Research Center poll found that 25% of US adults 50 to 64 and 10% of those 65 and older had used ChatGPT, OpenAI's flagship AI tool and the most popular chatbot. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) The Michigan survey, a telephone poll of 2,883 adults ages 50 to 97 across the US, looked not only at text-based chatbots like ChatGPT but also voice assistants like Amazon's Alexa and Apple's Siri. Most significantly, the survey found nearly all older adults want to be sure when they're receiving information generated by AI. "AI is here to stay. Many older adults seem to know about its benefits, yet most want more information about potential risks when using AI technologies," Robin Brewer, an assistant professor in the U-M School of Information, said in a statement. "The near-universal interest in clear labeling of AI-generated information should also be heeded by policymakers and the AI industry." The Michigan researchers found 51% of Americans aged 50 or older had used voice assistants in the past year, with 80% of those saying the devices helped them live independently and safely at home. AI-powered home security devices, like smart locks, cameras and alarm systems, were used by 35% of the adults, with 96% saying they helped them live independently. "Smart home devices frequently suggest using their latest features to watch over older relatives or help them live better lives," CNET smart home expert Tyler Lacoma said. "AI face recognition, for example, can send alerts to a family member if it spots someone leaving their home or granny pod to go shopping. Smart locks, meanwhile, make sure that older users never forget to lock the doors when they leave. I'm curious to see what the newest AI voice assistants like Gemini and Alexa Plus will bring to the table for older adults, perhaps by making more complex home technology easier to use with simple conversation." Another 14% of adults reported using AI to receive health-related information, although 47% of them said they'd rather interact with a person or by phone for that kind of information. Just 35% of the older adults surveyed said they'd be interested in using AI in their day-to-day lives, and while 58% expressed interest in knowing more about the benefits of AI, even more, at 81%, said they wanted to know more about the risks. Just over half of them said they thought AI would do more harm than good. That split generally aligns with the results of a broader Pew survey from earlier this year, which found 51% of Americans were more concerned than excited about AI. Read more: AI Essentials: 29 Ways You Can Make Gen AI Work for You, According to Our Experts The older adults in the Michigan survey also expressed understandable skepticism about the information generated by AI models, with 47% saying they had little or no trust in it. A similar portion said they weren't confident they could detect when it's incorrect. That problem is significant, especially as more information-gathering tools like search engines are built primarily on AI. Experts advise that you should always double-check information before making an important decision. "This inability to tell verified information from false information, whether in the form of text, image, video or audio, is especially important when it comes to health information," poll director Dr. Jeffrey Kullgren said in a statement.

Older Adults and AI: Poll Suggests a Wary Welcome | Newswise

Newswise -- The artificial intelligence revolution hasn't passed older adults by, a new poll suggests. But the data show that people over 50 have both curiosity and skepticism about AI, depending on how much they've used it, the data show. No matter what their level of experience with AI, nearly all people over 50 (92%) say they want to know when the information they read, see or hear is AI-generated, according to the new findings from the University of Michigan National Poll on Healthy Aging. In all, just over half (55%) of people age 50 and older have ever used an AI technology that they spoke or typed messages to for a variety of purposes on a variety of devices. Of those who used voice assistants like Amazon Alexa, Apple's Siri, or Google Nest in the past year, 80% said these devices are beneficial for helping them live independently and safely in their home, including 28% who said they're very beneficial for this purpose. In addition, 35% have used AI-enhanced home security devices; nearly all who had done so said they found them beneficial for helping them live independently and safely in their home. Overall, 14% of older adults polled said they have used AI technologies to receive health-related information. Among those who had, nearly half (47%) said a human interaction in person or by phone would be better for that purpose, with 26% saying AI and human interaction are about the same. The poll is based at the U-M Institute for Healthcare Policy and Innovation, and supported by Michigan Medicine, U-M's academic medical center. The poll team also analyzed data from adults age 50 and older in Michigan, with support from the Michigan Health Endowment Fund; those findings are at https://michmed.org/KDBJR and an interactive data dashboard is at michmed.org/GQkmw. "AI is here to stay. Many older adults seem to know about its benefits, yet most want more information about potential risks when using AI technologies," said Robin Brewer, Ph.D., M.S., an assistant professor in the U-M School of Information and a health AI researcher who worked with the poll team on the report. "The near-universal interest in clear labeling of AI-generated information should also be heeded by policymakers and the AI industry." She notes that many state legislatures have passed bipartisan bills to regulate AI use in recent years, including in political advertising and mental health care. In addition to asking about their AI experiences to date, the poll also asked older adults about their interest in using AI and learning about its risks, their trust level in AI-generated information, and their confidence in spotting incorrect information generated by AI. Just over a third of adults aged 50 and older (35%) said they're interested in using AI in their day-to-day lives. Over half (58%) would like to know more about the potential benefits of using AI, but a much greater percentage (81%) said they want to know more about the risks. Older adults are split on whether they think AI will do more harm or good, with 53% saying more harm than good and the rest saying the opposite. Nearly half of those polled said they had some trust in AI-generated information (49%) but only 4% said they had a lot of trust in it, and the rest (47% said they had little or no trust. Those who had used AI were more likely to express some level of trust. Also, half of older adults said they weren't confident they could detect when information generated by AI is incorrect. Prominent examples include "hallucinations", when an AI application makes something up that sounds correct but isn't true, or the "deepfake" videos and images that have proliferated in the past year. Scammers using AI-driven voice technology have also been known to attempt to swindle older adults over the phone. "This inability to tell verified information from false information, whether in the form of text, image, video or audio, is especially important when it comes to health information," said poll director Jeffrey Kullgren, M.D., M.P.H., M.S., a primary care physician at the VA Ann Arbor Healthcare System and associate professor of internal medicine at U-M. "Our new data show that people who report that their physical health or mental health are fair or poor were more likely to lack confidence in their ability to spot incorrect AI-generated information and were less likely to trust AI-generated health information, than those in good to excellent health," he said. "Clinicians, aging-focused organizations and health care organizations should all take notice." More outreach and education efforts by those organizations and government agencies could help, said Brewer, who studies how technology can be designed to support the needs of older adults and people with vision issues. "These findings present opportunities for older adults and their families to learn about 'healthy' AI use, including when to use AI instead of a search engine, or how to use AI to support healthy aging, such as accessing non-urgent health information," she said. The University of Michigan's Michigan Online platform offers free and paid online learning opportunities about artificial intelligence, including free mini-courses about generative AI and ChatGPT. More online learning about AI is in development; anyone can sign up to be notified of updates. Learn more about healthcare AI research and innovation by U-M faculty who are members of IHPI.