Across the country, researchers are working on numerous artificial intelligence-based research projects that could improve care and health outcomes for older Americans. Some technologies include robotic \u201cexosuits\u201d that help seniors stand and walk, smart home cameras that can detect movement and falls, and wearable devices \u2014 including in shoes \u2014 to monitor gait and balance.\u00a0<\/p>\n
The program is funded by the National Institute on Aging, a part of the National Institutes of Health, through a program called the Artificial Intelligence and Technology Collaboratories for Aging Research<\/a> (AITC). The AITC earmarked $40 million to fund promising AI-driven AgeTech projects, including tools for social isolation and loneliness, and algorithms to predict certain health outcomes or help with early diagnosis. For health care reporters, the initiative offers opportunities to examine how AI technologies aimed at older adults are moving from federally funded pilot projects into hospitals, homes and caregiving settings nationwide.<\/p>\n The NIA funded three AITC centers \u2014 at Johns Hopkins, the University of Pennsylvania and the University of Massachusetts Amherst<\/a>. Together, the three centers are part of the a2 Collective<\/a> that holds an annual symposium and webinars on the work and funds promising pilot studies nationwide.<\/p>\n \u201cThe NIA was at the forefront way before the hype about generative AI and large language models,\u201d geriatrician Peter Abadir, M.D.<\/a>, an associate professor of medicine at the Johns Hopkins University School of Medicine and co-principal investigator of the Johns Hopkins Artificial Intelligence and Technology Collaboratory for Aging Research<\/a> (JH AITC) in Baltimore, said in an interview.<\/p>\n Smart technologies for independent living<\/p>\n Smart homes, or embedding sensors or motion detection technology in homes, is one big area of focus, said George Demeris, Ph.D.<\/a>, director of the Penn Artificial Intelligence Collaboratory for Healthy Aging<\/a> (PennAITech) at the University of Pennsylvania, in an interview. This includes bed sensors to provide information on sleep patterns and restlessness, or door and window sensors that detect when people come and go. Observing how many times a person gets up to use the bathroom at night could indicate a urinary tract infection, he said, and help clinicians intervene before hospitalization is needed.<\/p>\n How AI could spot health risks earlier<\/p>\n \u201cRight now, our health care system is somewhat reactive, so we\u2019re waiting for something bad to happen, and a person to fall or experience another event, and then we try to minimize the impact of that,\u201d he said. In a more proactive approach, smart technologies could help identify changes in behavior or movement so interventions could happen sooner. Depth sensors, for example, provide a silhouette of the person as they move. By studying stride length, gait speed and balance, researchers could calculate a fall risk score, Demeris said. They also can see where seniors are leaning on unstable furniture to help them rise up from a chair.<\/p>\n \u201cI think we will have some form of AI in every home,\u201d he said.<\/p>\n Busting stereotypes about seniors using technology<\/p>\n There is a big myth that older adults will not use technology, said Abadir and Demeris. The Johns Hopkins AITC has a registry of 900 older adults who like to learn about what new tools are being tested, Abadir said. \u201cThey always want to play with some of those technologies, so they are open to trying and learning more,\u201d he said.<\/p>\n Older adults they have spoken with say they want independence, and to stay home and stay sharp, Abadir said. They want tools to help them perform daily tasks at home. The technologies don\u2019t have to be complicated, he said, even helping with simple things like opening a jar.\u00a0<\/p>\n This is especially important for people living in rural areas, according to Jason Karlawish, M.D.<\/a>, professor of medicine at the University of Pennsylvania Perelman School of Medicine and director of the Penn Memory Center. \u201cTechnologies offer a lot of opportunity and promise, removing barriers of travel and distance, getting people access to the right clinicians, and monitoring for problems that are not currently well monitored for.\u201d Continuous monitoring through AI technology also offers opportunities to help a person understand whether they\u2019re experiencing cognitive issues, which could be helpful for early detection and diagnosis, he said.<\/p>\n Seniors get a say<\/p>\n The Johns Hopkins AITC has a council of older adults, physicians and insurance company representatives that weigh in on what matters most to seniors and reviews technologies in development, Abadir said. Another core within the center works with companies to take products that seem promising into a studio to work with engineers and some older adults to try them out and make them feasible, he said.<\/p>\n When it comes to wearables, older adults don\u2019t want to wear items that look like medical devices, Demeris said. \u201cPeople don\u2019t like to wear something unless it\u2019s a fancy Apple Watch that their grandson also wears,\u201d he said. But seniors are becoming more comfortable with technologies like motion sensors that are already are embedded in retirement communities, he said. \u201cPeople are more welcoming of passive sensing technologies when it\u2019s framed as technologies that will help you stay independent and promote wellness, versus technologies that are going to tell you when you deteriorate and you\u2019re not doing as well.\u201d<\/p>\n Design for the user is important in technologies, the experts said. One company that Johns Hopkins worked with developed a non-camera Wi-Fi system that scans the home every 10 seconds to detect movement. As a resident moves around, it disturbs the Wi-XFWIFIFi signal, so an algorithm can detect how fast the person gets around, if they fall, their heart rate and breathing, Abadir said. It can send a warning to an emergency contact if adverse events such as falls occur. \u201cWhy is that a cool technology that we funded with a pilot? Because a lot of older adults don\u2019t want cameras but want to stay home safe,\u201d he said.\u00a0<\/p>\n By contrast, they trialed a headband designed to enhance sleep. But the interface it had was so complicated that seniors didn\u2019t want to use it, so they provided that feedback to the manufacturer.\u00a0<\/p>\n One of the biggest challenges in helping older adults adapt to AI-enabled tech is the lack of a qualified workforce to help people select, use and set up technologies, Karlawish said.\u201dThere\u2019s a real role for a profession that will help people pick and use and set things up. And not just a tech geek, but a brain geek, who\u2019s going to address what you need.\u201d<\/p>\n Unfortunately, there\u2019s a real gap in some of the information AI devices collect and what actually makes it to the physician, he added. \u201cIt\u2019s like a firewall between everything that could be monitored and what makes it into someone\u2019s electronic medical record. It\u2019s still something that needs to be addressed,\u201d according to Karlawish. \u201cThe problem isn\u2019t enough data. The problem is integrating it, culling it, sorting it and organizing it.\u201d<\/p>\n Localizing the story<\/p>\n Journalists could find multiple story ideas from the research ongoing at these centers, and can find ways to localize the story. Although the three centers are in Boston, Baltimore and Philadelphia, they are funding multiple pilot projects at other universities across the country. Johns Hopkins has funded research projects on AI technologies across 45 states, Puerto Rico and the Virgin Islands, according to a recent article<\/a> in the Johns Hopkins HUB publication. As of December 2025, the projects have resulted in 42 published research papers and brought seven new products to market, the article said. Some 45% of pilots are being done in startup companies, nearly half of which are women-owned, Abadir said.<\/p>\n