Discussion reveals challenges of AI-driven interventions in gerontological nursing, suggesting critical evaluation is necessary for effective implementation.
In an era where concerns about social isolation, discord, and our planet are rife, the coverage that artificial intelligence (AI) is receiving within and beyond healthcare is unprecedented. Publicity about AI far outstrips most people's basic familiarity with the crises of loneliness, conflict, and planetary health. Indeed, many conversations about AI in care for older people and nursing research that I hear seem oblivious to principles of effective health technology evaluation. The inescapability of AI today drives me back to evaluating this technological domain, prompting contemplation of disadvantages as well as advantages, along with accessibility, cost, and unintended effects. As a gerontological nurse, I always frame analysis of any health technology within the multidimensional process of ageing and with consideration of the needs and preferences of older people. AI offers a notable case in point as we gerontological nurses contend with evaluating this technology that is taking the healthcare industry and the scientific world by a paradoxical electric storm. Remember that AI represents a wide range of computerised functions and tools that are aimed at partially or completely replacing human actions. The word intelligence, however, is a misnomer. Machines are not intelligent. The allusion of intelligence comes with humans designing these tools in ways that mimic human function but can rarely even partially replace the higher-level aspects of our capacities. News coverage and social media posts constantly alert us to AI's manifold possibilities. Most excitement today about AI entails those tools that generate written or other content from data that already exist. As you might expect, this type of AI is called generative AI or GAI for short. Many of us are also aware of avid interest in integrating AI into robotics, especially for clinical applications that promote self-care among older people or extend nursing care for them. The AI domain is broad and diverse, garnering lots of enthusiasm but often spare critical analysis contextualised within our discipline. Activities involving AI are increasingly common across gerontological nursing research, practice, and education. Within the domain of gerontechnology, for example, many gerontological nurses are investigating the effects of AI-driven interventions like wearable devices or chatbots and other apps on functional outcomes and quality of life metrics. In addition to studying AI in relation to older people, researchers across healthcare are now adopting AI in research data management and analysis. Computer-assisted qualitative data analysis software, for instance, now commonly features AI assistance, which promises to generate findings for investigators more quickly than they could do themselves. Clinical uses of AI are increasingly frequent, too. Many gerontological nurses and other clinicians are recommending smart devices and AI-powered apps to older people and their care partners. Lastly, AI is very visible in nursing education, influencing the preparation of future gerontological nurses with everything from AI-supported editing of their assignments to AI-enhanced clinical simulation. Is there a revolution happening before our eyes? As I step back from coverage in the lay and professional media, several aspects of AI impress me in connection with ageing, older people, and gerontological nursing. Critically, intelligence in the term AI incorrectly implies the capacity for critical appraisal and evaluation. A machine or an algorithm cannot perform these actions. AI is a tool and, like all tools, it has strengths and limitations. Today, with enough money in our coffers and a bit of training at hand, we have a buffet of so-called smart AI tools to test out. From AI features in computer-assisted qualitative research data management and analysis platforms to smart apparel that stretches from head—smart spectacles—to toe—smart shoes, the options feel limitless. All these tools are touted as valuable to older people, nursing, and the study of ageing. But, as with all tools, successful use of AI relies on understanding relevant features and then precisely and accurately deploying those tools. Intelligence is human; artificiality is inherent in all AI tools. As AI has come to dominate healthcare and health research discussions, a different story deserves our attention. Human beings and their relationships are under remarkable stress because of a multiplicity of forces. Rapidly changing lifeways, geographic migration, economic shifts, human conflict, and the planetary crisis each contribute to phenomena that are deeply concerning to us as gerontological nurses and as global citizens. For instance, AI is repeatedly being offered as a solution for loneliness felt by older people, isolation and overburden experienced by care partners of older people, and overwork that nurses and our entire health and social care workforces live with every day. Too few people are thinking about the limitations inherent in interactions between human beings and chatbots or other AI-powered objects. Those interactions are merely that—interaction between human being and machine. In a discipline anchored by relationships, recognising the difference between interaction and an emotionally meaningful connection seems to recede into the background as we marvel at the potential of AI applications. Human beings, like most species, are fundamentally social. We cannot exist without social relationships. Many tools, like the automata of old (https://themadmuseum.co.uk/history-of-automata/) have engaged and entertained people for centuries. A machine mirroring the motions of a human interaction holds untold fascination for us. Computer technology and machine learning add a veneer that makes today's AI automata seem even more entrancing than in centuries past. Despite being entranced and entertained, we cannot forget that what we are seeking as human beings is relationships and not simply a sequence of interactions. At least for now, there are consistent and frequently predictable 'tells' of AI generated outputs including overuse of specific words as well as disconsonant content (Kobak et al. 2025). These 'tells' alert us to what is lacking. What makes relationships delightful is the unpredictability of other beings' responses and the rewards for participants in coming to know each other. Human beings recognise this quality in other humans and in other species. Just ask the owner of a companion animal—be it a dog, a cat, a rabbit, or a snake—about their pet's personality. Research on AI companionship has already begun to explore the possibility that long-term interaction with AI may alter brain function (Kosmyna et al. 2025). Such research suggests further questions about how using AI might alter our brain health and capacity for relationships. As an illustration, few nurses are exploring cognitive impacts—a withering of imagination and capacity for abstract thought, for example—in relation to repeated AI use. As gerontological nurses, we hold a distinctive vantage point on cognition, emotion, and related capacities. So far, there is no clear evidence and gaps in our understanding persist, underscoring a need for our scepticism and reflection on AI's role in healthy living and brain health in later life. While the remarkable usability of AI is oft touted, the environmental impacts are much debated. AI offers enormous potential as a tool for mitigating the climate crisis. Nonetheless, its growing use—particularly GAI—with current infrastructure represents a massive environmental threat with proportionately high use of water, electricity, rare elements and minerals, and resultant electronic waste (Bashir et al. 2024). Claiming the right to use AI in research or projects to serve human need without a larger viewpoint on the environment simply avoids addressing the delicate balance of our relationship with the earth. Efforts to exempt our own use of resources and engagement with industries that contribute to the climate crisis only perpetuate damages while curtailing potential benefit. Our deliberative, responsible use of AI and our activism to push for mitigating planetary harm are essential to limiting damage to the planet and all lifeforms. Importantly, we gerontological nurses need to think about the existential import of AI for phenomena within our domain. I began this essay by noting that we live in a time of loneliness, conflict, and planetary crisis. As nurses, we must take a broader view. For example, we can reflect on whether investigating AI-driven interventions for loneliness offers more sustainable solutions than working with and in communities to bring people together for better health and wellbeing. Whatever AI offers, it can only ever be an adjunct to what is innately human. Like other tools, AI may support our intelligence and relationships. It cannot replace them. People yearn for meaningful connections with other people, other beings like companion animals, and with nature. That interest in research that helps foster social connections and explores phenomena like animal-assisted therapy as well as investigating the health benefits of pet ownership and what is gained when spending time in nature pales in comparison with the buzz around AI should concern us all. Imagining that AI companions or chatbots will significantly remedy older people's unmet needs for social contact ignores the importance of relationships and, in fact, represents shockingly ableist thinking. The machine tolerance for repetitive interactions and responses free of judgement is often heralded as valuable for older people and especially those living with cognitive impairment. The stereotype of the garrulous older person unable to resist repeating the same stories has emerged clearly in promoting the shiny allure of chatbots immune to the tedium of listening to such repetitiveness. Yet many younger people also actively seek out AI social companion tools, sharpening the point of the ageist stereotype in a way that should draw our activism and shape our research agenda. I return now to the question that I posed earlier about AI constituting a revolution. AI represents a revolution only in terms of technology. For instance, the AI promises of savings in cost and time, ease, and higher volume or speed in targeted tasks are real only if its outputs are consistent and accurate. The use of smart devices can save time but only if data and alerts are accessible and comprehensible to all. Like all tools and interventions that we might consider using, AI promises benefits but may not warn of risks. Having more tools does not change the fundamentals of what it is to be human but the use of them does alter our productivity and may shape our functional capacities. This balance is not then a revolution for humankind. In the same way that we would never use any intervention indiscriminately, we cannot think of AI as a panacea for what does not work in healthcare or as today's on-trend solution. Its application requires our meticulous appraisal and testing before widespread adoption. The list of considerations for use of AI is long; the list of topics involving AI that need research is far longer. Our way forward is to hold fast to our principles, to explore the potential of AI, and to place that exploration in the larger perspective of what it is to be human and to need human relationships across the arc of a lifetime. The time is right to expand our research agenda in AI, framing it in terms that allow us to better address the crises of social isolation, conflict, and planetary health. With a shared perspective that AI represents but another set of tools—albeit a potentially remarkably powerful one—that we must use wisely, a great deal of work to effectively integrate AI into our research, education, and care lies ahead. Here at IJOPN, we look forward to reading your manuscripts reporting research that uses and examines AI. Meanwhile, please share your thoughts on AI in gerontological nursing and care for older people with us. IJOPN is on LinkedIn at https://uk.linkedin.com/in/international-journal-of-older-people-nursing-ijopn-10bb6674 and on Blue Sky at https://bsky.app/profile/intjnlopn.bsky.social. Just use our signature hashtag #GeroNurses as well as the hashtag #AIAndNursing when you tag us in your posts. No artificial intelligence was used in preparing this manuscript. The author declares no conflicts of interest. The author has nothing to report.
No takes yet. Share an insight, caveat, or question.
Sarah H. Kagan (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: