Older Americans Embrace AI for Independence & Safety, Study Shows
Artificial intelligence often dominates conversations in schools and workplaces, creating a perception that its adoption is primarily limited to younger demographics. However, a recent study challenges this notion, revealing that a significant number of older Americans are actively engaging with AI technologies. This raises crucial questions about their usage patterns, perceptions, and the implications for independent living.
A comprehensive survey conducted by a researcher specializing in older age, disability, and technology use, in partnership with the University of Michigan’s National Poll on Healthy Aging, delved into these very questions. The study polled nearly 3,000 Americans over the age of 50, exploring their AI usage, methods of interaction, and underlying concerns.
The findings indicate that 55% of the older adults surveyed had used some form of AI technology, whether through voice interaction, like Amazon’s Alexa, or text-based interfaces, such as OpenAI’s ChatGPT. Voice assistants proved overwhelmingly more popular; half of the respondents reported using one within the past year, compared to just one in four who had engaged with a chatbot.
The drive for independent living remains a paramount goal for many older Americans, often due to a desire to avoid or an inability to afford long-term care facilities. AI emerges as a promising tool to support this aspiration. The study shows that older adults who integrate AI into their homes find it genuinely helpful for maintaining independence and enhancing safety. While entertainment and information retrieval were common applications, some users demonstrated more creative uses, including generating text, creating images, or planning vacations.
Notably, nearly one in three older adults reported using AI-powered home security devices like smart doorbells, outdoor cameras, and alarm systems. An impressive 96% of these users felt safer as a result. While indoor monitoring cameras have sparked privacy debates, outward-facing cameras appear to offer a strong sense of security for those aging in place, particularly if they live alone or without nearby family.
However, the survey also highlighted that AI adoption among older adults is not uniform. Demographics play a significant role, with individuals in better health, possessing higher levels of education, and earning greater incomes being more likely to have used AI-powered voice assistants and home security devices in the past year. This pattern mirrors the historical adoption trends observed with other technologies, such as smartphones.
The question of trust in AI-generated content also yielded divided opinions. The survey revealed that 54% of older Americans trust AI, while 46% do not. Unsurprisingly, those who expressed greater trust in AI were more inclined to have used AI technology recently. This issue is compounded by the fact that AI-generated content can sometimes appear accurate but be factually incorrect. Worryingly, only half of the older individuals surveyed felt confident in their ability to identify inaccuracies in AI-generated information. Predictably, more educated users were more likely to express this confidence, while those reporting lower levels of physical and mental health were less likely to trust AI content.
These findings underscore a familiar cycle in technology adoption, where healthier and more educated individuals often lead the way in embracing and understanding new technologies. This raises critical questions about how to effectively reach all older people with information about both the benefits and risks of AI. How can non-users be supported in learning more to make informed decisions? And how can institutions develop better training and awareness tools to prevent over-reliance on AI or its inappropriate use for major decisions without a full grasp of the associated risks?
The survey results offer potential starting points for developing AI literacy tools tailored for older adults. A striking nine out of ten older people expressed a desire to know when information has been generated by AI. While some AI-generated content, such as Google search’s AI snippets, is now labeled, and some states have adopted policies for disclosing AI content in political ads, these notices could be made more visible across other contexts, including non-political advertising and social media. Furthermore, nearly 80% of older people expressed a desire to learn more about AI risks—where it might go wrong and what steps to take.
Policymakers have a crucial role to play in enforcing AI notices that clearly signal AI-generated content. This is particularly vital at a time when the U.S. is considering revisions to its AI policies that could potentially remove language concerning risk, discrimination, and misinformation. Overall, the study suggests that AI has the potential to support healthy aging. Yet, both over-trust and mistrust of AI must be addressed through improved training tools and policies that enhance the visibility of potential risks.