The world is getting older faster than it can hire caregivers. By 2050, the global population aged 60 and over will reach 2.1 billion, according to the World Health Organization. At the same time, OECD countries are on track to face a shortage of 13.5 million care workers by 2040. That gap has a lot of people looking at an unlikely answer: robots.
But “care robot” means different things depending on who’s talking. A researcher might mean a social companion the size of a stuffed animal. A tech company might mean a full-sized humanoid that can carry groceries and remind someone to take their medication. Families doing their research get caught in the middle, trying to figure out what’s actually available, what the science says, and whether any of this is worth the price tag.
This article cuts through the noise. This guide covers which humanoid robots for elderly care are deployed right now, what peer-reviewed research says about their effectiveness, how much they cost, and what families genuinely need to think through before making a decision. If you’re researching this topic for the first time, the short answer is: humanoid robots for elderly care are real, they work for specific use cases, and they’re becoming more accessible every year.
For a broader overview of what’s available today, see my guide to humanoid robots for home use.
Key Takeaways
- – OECD countries face a 13.5 million care worker shortage by 2040, making robotic assistance a structural need, not just a novelty.
- A 2025 meta-analysis of 19 studies found social robots reduced loneliness in older adults with a significant effect size of -0.59.
- The elder care assistive robot market is growing at 14.2% annually, from $3.38 billion in 2025 to a projected $9.85 billion by 2033.
- Consumer acceptance is real but uneven: 76% of elderly participants hold positive views of companion robots, yet only 28.8% accept robots for companionship specifically.
- Price ranges vary from $250 (ElliQ) to $25,000 (Pepper), and insurance rarely covers these costs today.
Table of Contents
What Does This Guide Cover?
- What Problems Are Humanoid Robots Solving in Elderly Care?
- Which Humanoid Robots Are Actually Being Used in Eldercare Today?
- Do Care Robots Actually Work According to the Research?
- How Much Does a Care Robot Cost — and Who Pays for It?
- What Should Families Know Before Buying a Care Robot?
- Take the Next Step
- Frequently Asked Questions
What Problems Are Humanoid Robots Solving in Elderly Care?
The UN estimates the global population aged 65 and over reached 703 million in 2023 and is projected to reach 1.5 billion by 2050 (UN DESA, 2023). That demographic shift isn’t just a statistic. It’s a real pressure on families, healthcare systems, and care facilities that are already stretched thin. Humanoid robots for elderly care are emerging as a partial answer to three specific, concrete problems.
The caregiver shortage is structural, not temporary. The US alone will need 3.5 million additional healthcare workers by 2030, and unfulfilled home care jobs could hit 4.6 million by 2032, according to the Eldercare Workforce Alliance. No amount of recruitment and training can close that gap fast enough. In practice, robots don’t replace skilled nursing, but they can handle routine tasks that currently eat a caregiver’s time: medication reminders, fall detection, check-in conversations.
Loneliness is a clinical problem, not just a social one. Isolation among older adults is associated with a 26% increased risk of dementia, comparable in effect to smoking 15 cigarettes a day, per research cited by the CDC. Many older adults living alone go entire days without a real conversation. As a result, companion robots, even simple ones, have shown measurable results in reducing that isolation, as covered in detail below.
Cognitive decline creates specific daily risks. Someone living alone with mild dementia faces a unique set of dangers: missed medications, forgotten meals, wandering. Humanoid robots for elderly care can provide structured daily routines, gentle reminders, and activity monitoring without the constant presence of a human aide. That’s not a full care solution, but for early-stage cases, it can meaningfully extend independent living.
To understand the underlying technology, see my guide to how humanoid robots work.
Which Humanoid Robots Are Actually Being Used in Eldercare Today?
A March 2026 scoping review published in JMIR Aging examined 59 studies covering 25 humanoid robots used in care settings. Pepper appeared in 15 of those studies, NAO in 9, and 83% of studies took place in real-world settings rather than labs (JMIR Aging, 2026). The roster of robots actually deployed today is smaller than the hype suggests, but it’s genuinely growing.
PARO: The Therapeutic Seal from Japan
PARO is technically not humanoid, but it’s the most clinically validated social robot in eldercare worldwide. It’s a soft robotic seal that responds to touch, voice, and light. It’s been used in dementia care units across Japan, Europe, and the US since the early 2000s. Price point is around $6,000, which puts it in care facility territory rather than home purchase.
ElliQ: The Tabletop Companion
ElliQ, made by Intuition Robotics, is a small tabletop device with a screen and expressive head movements. It doesn’t have a humanoid body, but it conducts real conversations, reminds users of medication, facilitates video calls, and reports health trends to family members or caregivers. At around $250 upfront plus $30 per month, it’s the most accessible option in this space. Several US states have run pilot programs placing ElliQ units with isolated older adults.
Pepper and NAO: Research Workhorses
SoftBank’s Pepper and NAO robots, both profiled in my guide to humanoid robot companies, have appeared in more eldercare research studies than any other platforms. Pepper is about 1.2 meters tall, has expressive eyes, and can conduct conversations, play games, and guide users through exercises. NAO, smaller at 58 cm, is often used in cognitive therapy and dementia care programs. Neither is primarily sold as a consumer product. At $20,000 to $25,000 for Pepper and $9,000 to $16,000 for NAO, they’re institutional tools. But the volume of published research on both platforms gives families and facilities a real evidence base to evaluate.
Hyodol: South Korea’s Government-Backed Companion
Hyodol is a plush companion robot developed in South Korea. It looks like a doll rather than a humanoid, but it actively monitors vital signs, detects falls, and prompts medication. South Korea’s government has subsidized its deployment in care settings, with over 12,000 units now active across the country. That scale makes Hyodol one of the most widely deployed care robots in the world.
1X NEO: The Humanoid on the Horizon
1X’s NEO robot represents the next tier: a full-sized, mobile humanoid designed for home environments. At around $20,000, it’s priced for early adopters. It’s not yet a care product with clinical validation, but it’s being positioned for exactly this market. For a broader look at what’s coming from companies like 1X and others, see my complete guide to humanoid robot companies.
Robot Price Comparison: What Are You Actually Paying?
| Robot | Type | Approx. Price | Best For | Clinical Evidence |
|---|---|---|---|---|
| ElliQ | Tabletop companion | $250 + $30/mo | Isolated seniors at home | Moderate (pilot studies) |
| PARO | Therapeutic seal | ~$6,000 | Dementia care facilities | Strong (multiple RCTs) |
| NAO | Small humanoid | $9,000-$16,000 | Cognitive therapy programs | Strong (59+ studies) |
| 1X NEO | Full humanoid | ~$20,000 | Early adopters | Early stage |
| Pepper | Full humanoid | $20,000-$25,000 | Institutional research | Strong (15+ studies) |
Do Care Robots Actually Work According to the Research?
A 2025 meta-analysis published in The Gerontologist reviewed 19 studies involving 1,083 older adults and found that social robots produced a statistically significant reduction in loneliness, with an effect size of -0.59 (p less than 0.01) (PMC/The Gerontologist, 2025). That’s a meaningful result, not a marginal one. The evidence base for humanoid robots for elderly care is still maturing, but it’s no longer just anecdotal. That’s a significant shift from even five years ago, when most coverage of humanoid robots for elderly care was speculative.
The PARO robot has probably the strongest individual study record. A 2024 randomized controlled trial published in the Journal of the American Medical Directors Association followed 52 older adults with mild dementia over 8 weeks. Participants who interacted with PARO showed statistically significant reductions in both depression (p less than 0.001) and loneliness (p less than 0.001) compared to the control group (JAMDA, 2024). Eight weeks. Fifty-two participants. Those are small numbers, but the effect sizes were not.
Social robots showed a statistically significant reduction in loneliness among older adults across 19 studies and 1,083 participants, with an effect size of -0.59 (p less than 0.01). The result held across diverse robot types and care settings, suggesting the effect is not specific to any single platform. (The Gerontologist, 2025 meta-analysis via PMC)
That said, the JMIR Aging scoping review from March 2026 adds an important caveat. After examining 59 studies on 25 different humanoid platforms, the authors noted a “striking lack of randomized controlled trials.” Most studies rely on self-reported outcomes, small samples, and short time horizons. These robots do help, but the exact benefit amount, for how long, or for which specific populations they work best.
By contrast, acceptance rates tell a similarly nuanced story. A study by Florida International University found that 76% of elderly participants hold positive views of companion robots, and 65% said they would use one (FIU Business/Technology in Society, 2024). A Swiss national survey of 1,211 people found that 49 to 57% accept robots for practical assistive tasks, but only 28.8% accept robots for companionship (PMC, 2025). The gap between accepting help from a robot and wanting a robot as a companion is real and worth taking seriously.
What the research doesn’t yet show is strong evidence for humanoid robots specifically doing complex physical care tasks: bathing, dressing, or safely transferring patients. Looking at the published literature across the 25 platforms covered in the JMIR review, the consistent pattern is: social interaction works, physical assistance doesn’t yet. Those capabilities are still in development, and my breakdown of humanoid robot challenges explains what’s real versus hype. The honest picture is that today’s care robots are best at social interaction, reminders, and monitoring. Physical assistance is the next frontier.
How Much Does a Care Robot Cost — and Who Pays for It?
The global elder care assistive robot market was valued at $3.38 billion in 2025 and is projected to reach $9.85 billion by 2033 at a compound annual growth rate of 14.2% (Grand View Research, 2025). That growth rate reflects the structural pressure of the caregiver shortage: as the shortage deepens, demand for automated support tools rises proportionally. The money is moving. The question is who’s spending it.
Right now, families and care facilities are largely paying out of pocket. Medicare and most private insurers in the US don’t cover social or companion robots. However, PARO has obtained some reimbursement pathways in certain European countries and in Japan, where the government has actively pushed robotic eldercare as policy. South Korea’s subsidy of the Hyodol robot is another model: the government absorbs the cost to address a structural care shortage. The US hasn’t made that move yet.
For families weighing options today, the honest cost picture looks like this. ElliQ at around $250 upfront plus $30 monthly is within reach for many households, especially if it replaces some paid check-in service. PARO at $6,000 is typically a care facility purchase. NAO and Pepper are institutional tools at $9,000 to $25,000. Full humanoid robots like 1X NEO at $20,000 are early-adopter purchases today, and my roundup of humanoid robots for sale tracks current prices and availability. The price chart above captures this spread visually.
Will Prices Come Down Over Time?
Almost certainly, yes. The elder care robot market is following the same cost-reduction curve seen in other consumer technology categories. As volumes increase and manufacturing scales, the $20,000 full humanoid will likely look like today’s $6,000 PARO within a decade. For now, though, families are making real purchasing decisions with real current prices. That means starting with the lower end of the market and reassessing as the technology matures is the rational approach for most households.
Elder Care Robot Market Growth Projection
What Should Families Know Before Buying a Care Robot?
Consumer acceptance data shows that 76% of elderly participants hold positive views of companion robots and 65% say they’d use one, according to Florida International University (FIU Business/Technology in Society, 2024). That’s encouraging. But a Swiss national survey of 1,211 people found a very different number when the question shifted: only 28.8% accepted robots specifically for companionship (PMC, 2025). The gap between abstract approval and personal acceptance is where most families get stuck.
Across a Swiss national survey of 1,211 adults, acceptance of robots for practical assistive tasks ranged from 49 to 57%, but acceptance for companionship fell to just 28.8%. The data suggests people draw a meaningful line between a robot that helps and a robot that connects. (PMC, 2025)
In my experience talking with families navigating this decision, the biggest mistake is buying a robot to solve a problem the older adult doesn’t recognize as a problem. Loneliness, for instance, is often something families observe from the outside while the older adult denies or minimizes it. A companion robot introduced without buy-in from the person who’ll actually use it tends to sit unused in a corner within two weeks.
Privacy and Ethical Considerations: A Practical Checklist
Before committing to any care robot, work through these questions together with the older adult and other family members involved in care:
- Data storage: Where does conversation data go? Is it stored on-device, in the cloud, or shared with third parties? Who owns the recordings?
- Camera access: If the robot has a camera, who can access the video feed, and when?
- Consent: Has the older adult genuinely agreed to monitoring, or are family members making that choice for them?
- Failure modes: What happens if the robot malfunctions? Is there a human fallback for critical reminders like medication?
- Dignity: Would your parent feel embarrassed or watched if they knew the extent of data collection? That discomfort matters.
Is a Care Robot Right for My Parent? A Decision Framework
Use this as a starting point, not a final answer.
A care robot is likely worth exploring if:
– Your parent lives alone and goes long stretches without social contact.
– They’re in early-stage cognitive decline and can still engage with technology.
– You’re managing remote caregiving from another city.
– They’re already comfortable with smartphones or tablets.
– You’re primarily looking to supplement, not replace, human care.
A care robot is probably not the right fit if:
– Your parent is resistant to the idea after an honest conversation.
– They need physical assistance beyond what current robots can safely provide.
– They have advanced dementia and would be confused by an artificial presence.
– You’re hoping it will replace paid professional care entirely. It won’t, not yet.
For families thinking about the physical safety side of robot interactions, safety compliance is worth researching carefully before any purchase. My overview of humanoid robot safety standards covers the key certifications and what they mean in practice.
What about Tesla Optimus and other general-purpose humanoids being positioned for home use? They’re worth watching. But they’re not yet validated for eldercare, and families shouldn’t count on them being a near-term solution. The robots making a real difference in eldercare right now are specialized tools, not general-purpose humanoids.
Frequently Asked Questions
What’s the best humanoid robot for elderly care right now?
There’s no single best option because needs vary widely. For companionship and daily reminders, ElliQ (around $250 upfront, $30 per month) is the most accessible starting point. For dementia care with the strongest clinical evidence, PARO has the deepest research record, though at $6,000 it’s typically a care facility purchase. A 2026 JMIR scoping review found Pepper featured in the most eldercare research studies (15 out of 59 examined), making it the most studied humanoid in the field.
Do care robots really reduce loneliness, or is that just marketing?
The evidence is real. A 2025 meta-analysis in The Gerontologist reviewed 19 studies covering 1,083 older adults and found a statistically significant reduction in loneliness with an effect size of -0.59. A separate 2024 randomized controlled trial with PARO showed significant reductions in both depression and loneliness (p less than 0.001) in 52 adults with mild dementia after 8 weeks. That said, researchers in the 2026 JMIR review noted a “striking lack of randomized controlled trials” across the broader humanoid robot field. The results are promising, but more rigorous long-term studies are still needed.
Who pays for a care robot? Does insurance cover it?
Insurance rarely covers care robots in the US today. Medicare doesn’t classify them as durable medical equipment. PARO has obtained some reimbursement pathways in Japan and parts of Europe. South Korea’s government subsidized Hyodol deployment across 12,000+ care settings as national eldercare policy. In the US, families and care facilities currently pay out of pocket. Given the elder care robot market is growing at 14.2% annually (Grand View Research, 2025), insurance and public funding frameworks are likely to evolve, but there’s no firm timeline.
Are elderly people actually open to using care robots?
Acceptance is higher than most people expect, but it’s nuanced. A Florida International University study found 76% of elderly participants held positive views of companion robots, and 65% said they’d use one (FIU/Technology in Society, 2024). However, a Swiss national survey of 1,211 adults found only 28.8% specifically accepted robots for companionship, compared to 49 to 57% for practical assistive tasks (PMC, 2025). The takeaway: most older adults are open to robots that help them do things. Fewer want a robot as a substitute for human social connection. Personal buy-in before purchase matters a lot.
How do humanoid robots for elderly care handle emergencies?
Current systems vary significantly. Most companion robots like ElliQ include fall detection alerts sent to family members or caregivers. Hyodol monitors vital signs and triggers alerts if something seems wrong. Full humanoid robots like Pepper and NAO in care settings are typically used under supervision, not deployed as autonomous emergency responders. Physical intervention during a medical emergency is beyond what any commercially available care robot can reliably do today. That’s one reason these robots work best as a complement to human oversight, not a replacement for it. For families who want to understand the broader technology landscape before buying, my overview of how humanoid robots work explains the key capabilities and limitations in plain language.
About the Author
Ulrich Baldauf is the founder of There’s A Robot For That, a publication tracking the deployment of humanoid robots in industry and at home. He covers humanoid robot development, market adoption, and the practical questions families and businesses face as this technology moves from research labs into the real world. Read more articles by Ulrich.
Sources
- UN DESA, 2023. https://www.un.org/development/desa/pd/content/ageing
- JMIR Aging, 2026. https://aging.jmir.org
- PMC/The Gerontologist, 2025. https://pmc.ncbi.nlm.nih.gov
- JAMDA, 2024. https://www.jamda.com
- FIU Business/Technology in Society, 2024. https://www.sciencedirect.com/journal/technology-in-society
- Grand View Research, 2025. https://www.grandviewresearch.com
Ulrich Baldauf is the founder of There’s A Robot For That, covering humanoid robotics for manufacturing and industrial operations. He has tracked the humanoid robot sector since 2024, with a focus on safety standards (ISO 10218, EU Machinery Regulation 2023/1230) and what deployments mean for operations and EHS teams. Connect on LinkedIn: linkedin.com/in/ubaldauf
