A 78-year-old widow in suburban Denver receives a small white robot named ElliQ for her birthday. It greets her by name in the morning, reminds her to take her blood pressure medication, asks how she slept, suggests she go for a walk, plays the music she loved in her thirties, and tells her she is loved. Her adult children, who live in three different states and visit on holidays, find that she seems brighter when they call her now. They are relieved. They are also vaguely uneasy. They are not sure why.
This page is about the unease.
AI is moving into elder care faster than the Catholic Church has had time to teach about it. Companion robots for lonely seniors, dementia tracking apps, algorithmic monitoring of aging parents, voice assistants that double as social presence, AI-driven recommendation systems in nursing homes. Some of this technology is genuinely useful. Some of it raises questions the Catholic tradition has been thinking about for two thousand years. This page offers a framework for sorting which is which.
The loneliness emergency, and what AI is trying to solve
Begin with the problem. Loneliness among older adults is real, measurable, and lethal. The U.S. Surgeon General's 2023 advisory on the loneliness epidemic found that social isolation in older adults is associated with risk of dementia, heart disease, stroke, and premature death at levels comparable to smoking. Roughly one in four adults over 65 in the United States is considered socially isolated. The numbers in Europe and Japan are similar or worse.
At the same time, the workforce of human caregivers is shrinking relative to the population that needs care. Birth rates have fallen. Adult children live further from aging parents than they did fifty years ago. Care workers are in short supply, and the supply is concentrated in regions wealthy enough to attract them. The math of who will care for the next generation of elderly is not encouraging.
Into this gap walks AI. Companion robots like ElliQ, designed for older adults living alone, are now being prescribed by insurance plans. AI-driven monitoring systems track movement, falls, sleep, and medication adherence in homes and care facilities. Voice assistants engage in small talk and report changes in mood to remote family members. AI tools designed for dementia care play personalized music, simulate familiar faces, and provide structured conversation. The marketing copy is compelling: dignity, autonomy, companionship, peace of mind.
None of this is wrong as a starting point. The need is real. The Catholic question is what the technology is doing inside the need, and what it might be doing to it.
The Catholic theology of aging
Before evaluating the technology, the framework. Catholic teaching on aging rests on a small number of clear commitments.
The dignity of the person is intrinsic and does not diminish. The aging person, the person with dementia, the person on the threshold of death, has exactly the same dignity as the healthy adult. This is doctrinally fixed and is not a sentimental claim. It is the same dignity claim the Church makes for the unborn and for the dying, and it is built on the same foundation: that the human person is made in the image of God, and that this image is not reducible to function, productivity, or cognition.
Dependency is not a defect. The Catholic tradition treats dependency as a feature of every human life, not a problem to be solved. Infants are dependent. The sick are dependent. The aging are dependent. So, at various points, are all of us. A society that treats dependency as a failure of personhood is a society that has misunderstood what persons are. Pope Francis returned to this theme repeatedly, calling the elderly a treasure of the Church and warning against what he called a throwaway culture that treats aging persons as burdens.
The Fourth Commandment binds across the life course. "Honor your father and your mother" has been read by the Catholic tradition as a lifelong commitment that intensifies, rather than weakens, as parents grow old. The Catechism states it directly: "as much as they can, they must give them material and moral support in old age and in times of illness, loneliness, or distress" (CCC 2218). The duty is not transferable to a machine. It is also not transferable to a paid caregiver, although a caregiver can support its discharge. The bond is personal and the obligation is personal.
Care is relational, not transactional. Catholic moral theology has always treated care as a form of relationship rather than a form of service delivery. The person being cared for is encountered as a person, not as a problem to be managed. The carer is changed by the act of caring, not merely an instrument of it. This is why the Church has historically resisted any framing of care that reduces it to a contract between a provider and a consumer.
With those four commitments in place, the question of AI in elder care becomes tractable.
What the Church has already said about AI in care
The 2025 Vatican doctrinal note Antiqua et Nova, on the relationship between artificial and human intelligence, addressed AI in care relationships directly. The relevant passages are blunt.
Antiqua et Nova warned that AI's capacity to simulate emotional response and relational presence is being used in ways that take advantage of human vulnerability. The document called the use of AI to deceive in education or in relationships, including human sexuality, "immoral" and requiring "careful vigilance." Although the note did not name elder care specifically, the principle extends with full force to the use of AI companion technology with vulnerable older adults, particularly those whose cognitive state limits their capacity to evaluate what is and is not a real relationship.
Pope Leo XIV's first encyclical, Magnifica Humanitas, releasing May 25, 2026, is expected to develop this teaching further. Pope Leo XIV's 2026 World Communications Day message put the theme plainly: AI's simulation of human voices and faces "encroaches upon the deepest level of communication, that of human relationships." When the relationship in question is a daughter's relationship to her aging mother, or a son's to his father, the encroachment is not abstract.
The Church's position is not against AI assistance. It is against AI as a sustainable replacement for the human encounter that elder care requires.
The distinction that matters: supplement vs. substitute
The single most useful distinction in this whole field is between AI as a supplement to human care and AI as a substitute for it. The Catholic framework approves the first and is deeply concerned about the second. Here is how the distinction plays out across the most common technologies.
Safety monitoring. Fall detection sensors, medication reminders, alerts for unusual sleep patterns, AI-driven analysis of voice or movement that flags potential health changes. These tools extend the care a family or human caregiver provides. They give a remote daughter the ability to know her father has not fallen overnight, which is a service to the relationship rather than a substitute for it. Catholic teaching has no general objection to these tools, although the questions of consent and dignity that surround any monitoring of an adult apply. The aging parent is still a person with the right to know they are being watched and the right to refuse it.
Practical assistance. AI-driven scheduling, voice assistants that help with daily tasks, automated reminders, software that helps an older adult manage email or stay connected with family. This is assistive technology of the kind the Church has always supported. The Catholic concern is not present here.
Cognitive stimulation. AI applications that engage older adults in puzzles, conversations, music, or memory exercises, particularly in early dementia care. The evidence for benefit is reasonable. The Catholic question is whether the tool is being used alongside human contact or in place of it. Used alongside, it is unobjectionable and can be genuinely helpful. Used in place of, it raises the substitution concern.
Companion robots. This is where the framework gets tested. ElliQ, PARO the robotic seal used in dementia care, AI companion apps designed to provide ongoing conversation and emotional response. These tools are marketed as companionship, and there is research showing they reduce measured loneliness and agitation in some users. The Catholic concern is not that the tools are dangerous in isolation. It is that they are being deployed at scale into a society that has not solved the underlying problem, which is that older adults are alone because the people in their lives are not present. When a robot becomes the easy answer to that problem, the harder answer, which is human presence, becomes harder to ask for.
AI in dementia care. This is the hardest case in the whole field, and it deserves its own section.
The dementia case, where the framework is hardest
A person with advanced dementia may genuinely seem to enjoy interaction with a robotic seal that responds to touch with warmth and motion. Care researchers have measured reduced agitation, reduced use of antipsychotic medication, and apparent improvements in quality of life. The same person may not be able to remember whether their daughter visited yesterday, but may light up when the robot is brought to them. What is Catholic teaching supposed to say about that?
The honest answer is that the question is harder than the categorical responses in either direction would suggest.
The strict view, which some Catholic ethicists have argued, is that the interaction is a form of deception. The person believes they are in a relationship and they are not. The Catholic prohibition on deception, particularly of vulnerable persons, applies. The robot should not be used.
The pragmatic view, which other Catholic ethicists have argued, is that the deception framing misses what is actually happening. The person with advanced dementia is not making the kind of cognitive claim that the strict view assumes. They are not concluding that the seal loves them. They are responding to warmth, motion, and presence in a way that reduces their distress. Refusing them that reduction in distress on the grounds of philosophical purity is itself a failure of care.
The Catholic framework I would offer here, which I think both positions have to grapple with, is the following. The question is not whether the interaction is real to the person. It is whether the interaction is real to the universe. The robotic seal is not loving the person back. The AI companion is not present in the way that another person is present. And a society that increasingly meets the relational needs of its most vulnerable members with simulations is a society that has made a serious choice about what those members are owed. The dementia patient may not be able to register that choice. The choice is still being made about her.
None of this means a particular family should not use a tool that genuinely helps. It does mean the tool should not let the rest of us off the hook. The robotic seal that calms an agitated patient on a Tuesday afternoon is one thing. The widespread deployment of robotic seals as a way to avoid the harder conversation about how many human carers we are willing to fund is another. Catholic ethics has to be able to see both.
The Fourth Commandment in an AI economy
The hardest place this framework lands is on adult children of aging parents, and I write this section knowing that I am one of them and that the demands of the position are real.
"Honor your father and your mother" is the only commandment in the Decalogue with a promise attached. The Catholic tradition has read it as a lifelong obligation, intensifying rather than diminishing as parents age. The obligation is personal. It can be supported by paid caregivers, by extended family, by community, by parish. It cannot be discharged by a machine.
AI in elder care is being marketed, sometimes explicitly, as a way to ease the burden on adult children. The phrasing should be read carefully. The burden that the Fourth Commandment imposes is not, in the Catholic tradition, a burden to be eased. It is a duty to be borne. The bearing of it is part of what makes the person doing the bearing into the kind of person Catholic teaching wants them to be. An adult child who relies on AI to maintain the appearance of care while being substantively absent has not solved the Fourth Commandment. They have routed around it.
This is a hard thing to say in a world where the children are stretched, the distances are real, the working hours are long, and the alternative to AI assistance is often nothing. I am not pretending the situation is simple. But Catholic teaching is also not in the business of telling adult children that the duty has gotten easier because the tools have gotten better. The duty is what it has always been. The tools are useful where they extend the discharge of the duty. They are corrosive where they replace it.
What to look for, what to refuse
For an adult child or family caregiver evaluating an AI tool for an aging parent, the questions that surface the Catholic framework are concrete.
Does the tool make it easier to spend genuine time with the person, or easier to spend less time with the person? Does it report back to family members in a way that strengthens family bonds, or in a way that lets family members feel informed without being present? Does the parent know what the tool is doing and have they agreed to it? Does the tool respect the parent's autonomy and preferences, or is it being used to manage them? When the parent's cognitive state changes, does the family revisit the decision, or does the tool become the default?
For companion technology in particular: is the device a supplement to a rich relational life, or is it filling a vacuum that should be filled with people? If the device were taken away tomorrow, would anyone notice? If the answer is no one, the device is doing too much.
Refuse, in the Catholic framework: any tool that is being used as the primary relational presence in a vulnerable person's life. Any tool that is being deployed to disguise the absence of family or community. Any tool that operates without the knowledge or consent of the person it is operating on. Any tool whose purpose is to manage the person rather than to serve them.
Accept, in the Catholic framework: tools that extend the reach of human care. Tools that make safety, medication adherence, and practical assistance better than they otherwise would be. Tools that connect aging parents to family and friends. Tools used alongside, rather than instead of, the human presence that Catholic teaching insists on.
What the Church is still working out
This is not a closed question in Catholic moral theology. The technology is new and the magisterium has only begun to address it directly. Antiqua et Nova in January 2025 was the first Vatican document to engage AI in care relationships with any specificity, and even there elder care was not named. Pope Leo XIV's encyclical, Magnifica Humanitas, releasing May 25, 2026, is expected to address AI in human relationships more broadly, and the application to elder care will follow.
What can be said now is that the framework is in place, even if the specific applications are still being argued. The framework is the dignity of the human person, the Fourth Commandment, the irreducibility of relational care to information processing, and the warning against using technology to disguise the absence of love.
An AI tool for an aging parent is a tool. Like all tools, it can serve human flourishing or it can quietly displace it. The Catholic vocation is to use the tools that help, refuse the tools that harm, and never lose sight of the person at the center of the question, who has the same dignity at 85 that she had at 25, and who is owed exactly what she has always been owed: the people she loves, present in her life, for as long as that life lasts.