Guide 1
A clear account of which technologies are deployed in UK care settings, what they do day to day, and the distinction that most coverage gets wrong.
If you search for "robots in care homes", you will quickly encounter two very different things presented as if they are the same: highly capable ambulatory humanoid robots shown in demonstration videos, and the companion and communication devices that are actually operating in UK care homes. The gap between these two categories matters enormously for anyone trying to make a real decision.
This guide sets out what is actually deployed, what it does, and what it cannot do. It also explains where the more advanced humanoid technology currently stands and why it is not yet operational in UK care at any meaningful scale.
The most important thing to understand about robots in UK care is that the category covers at least two substantially different types of technology.
The first category is companion and communication robots: devices, often on wheels or with a screen-based interface, that provide social interaction, reminders, video communication with family, and AI-assisted conversation. These are genuinely operational in UK care settings today. They are useful, carefully evaluated in a number of documented pilots, and increasingly commercially available. They are not humanoid in the science-fiction sense; they do not walk, and they do not perform physical care tasks.
The second category is ambulatory humanoid robots: human-shaped machines capable of independent movement and, in research settings, basic physical manipulation. Companies in this space include Boston Dynamics and several well-funded US and European startups. These devices are impressive in controlled demonstrations. They are not deployed in UK care homes at scale. The technology, the regulatory pathway, the insurance and liability frameworks, and the evidence base required to deploy a walking robot in a care setting are all still in development. Anyone representing otherwise is overstating the current position.
This guide focuses on what is real and operational in the UK today. The homepage has more on the broader landscape.
The most documented category of robot currently active in UK care settings is the AI-assisted companion and communication device. These systems share several characteristics: they sit in communal areas or in residents' rooms, they can engage residents in conversation, they support video calls with family members, they can deliver reminders about medication or daily routines, and some include entertainment or activity functions.
Service Robotics has operated GenieConnect units in UK care homes. These devices appear as a screen-based interactive companion, capable of conducting conversations, supporting family video calls, providing medication reminders, and delivering news and entertainment content. Documented deployments have involved care homes in England, and the company has reported outcomes including reduced feelings of isolation among residents and reduced staff time spent on routine communication tasks.
GenieConnect is an example of what is actually in UK care settings today. It is a communication and companionship device, not a humanoid robot. It does not perform physical care tasks.
Cera, a UK care and technology company, has reported piloting AI-assisted communication droids in residential care settings in London. These devices have similar functions to GenieConnect: social interaction, family communication, and routine reminders. Cera has positioned this work as part of a broader technology-assisted care model designed to support care workers rather than replace them.
Again, these are companion and communication devices. Their value is in the interaction, conversation and connection they provide, not in any physical capability.
Across documented UK deployments, companion communication robots share a consistent pattern: they address social isolation and improve family connection, they support staff with routine communication tasks, they work best alongside human carers rather than in place of them, and their outcomes depend significantly on how they are introduced and supported within the care setting.
They also share consistent limitations. They cannot provide physical care of any kind. They cannot administer medication. They cannot support a resident with mobility, personal care, or falls prevention. They cannot substitute for the judgment and presence of a trained carer. In some resident profiles, particularly advanced dementia, the technology may not engage effectively at all. These are not failures of the technology; they are its honest boundaries.
It is worth being explicit, because the gap between expectation and reality is wide in this sector.
Robots currently deployed in UK care cannot: provide personal care (washing, dressing, continence care); support or assist with mobility or transfers; administer medication; prevent or respond to falls; substitute for regulated clinical or nursing judgment; or provide the kind of attentive, responsive human presence that many residents need most.
These limitations are not specific to any single product. They reflect the actual state of the technology, the regulatory environment, and the liability and insurance landscape. A care home that deploys a companion communication device in the expectation that it will reduce staffing requirements for personal care has misunderstood what the technology does.
This is not a reason not to deploy the technology. Reducing social isolation is genuinely valuable. Improving family connection is genuinely valuable. Freeing staff from routine communication tasks so they can focus on direct care is genuinely valuable. But the value is in what the technology actually does, not in what it has sometimes been sold as doing.
Ambulatory humanoid robots are real, and they are developing rapidly. Boston Dynamics' Atlas robot, Tesla's Optimus, and various research-stage systems can walk, navigate environments, and perform limited manipulation tasks in controlled conditions. Research programmes at UK universities, including work at the University of Bristol and at King's College London, have explored robotic systems in health and care contexts.
None of this translates to operational deployment in UK care homes. The path from laboratory demonstration to regulated care setting involves solving problems that go well beyond the mechanics of walking: safety certification in environments shared with vulnerable people, insurance and liability frameworks that do not yet exist in any robust form, clinical evidence standards for a new category of care intervention, and the practical realities of maintenance, failure modes, and 24-hour operational demands.
The UK social care sector does face a genuine workforce challenge. The demographic pressures are real. The interest in technology as part of the solution is understandable and legitimate. But the humanoid robot that takes on physically demanding care tasks is a medium-to-long-term prospect. Planning decisions made today should be based on what is available today.
The case for companion and communication robots in appropriate care settings is straightforward and rests on documented evidence. Social isolation is a serious problem in residential care. Family contact matters for wellbeing. Staff time spent on routine communications has an opportunity cost. Companion devices address all three, and documented pilots in the UK have reported positive resident responses in appropriate settings.
The case against, or the risks, are equally important to state plainly. There is a risk of over-reliance: a care home that uses companion technology as a substitute for adequate staffing, rather than as a supplement, is using it harmfully. There is a risk of unsuitable deployment: residents who find the technology confusing or distressing, or whose care profile means they cannot engage with it, should not have it imposed on them. There is a risk of inflated expectation: families or commissioners who expect the technology to do more than it does will be disappointed, and that disappointment can fall on the residents.
Managed well, with appropriate consent processes, staff involvement, and honest evaluation, companion robot technology can make a genuine difference in the right settings. The key phrase is "managed well". The technology does not manage itself.
If you are a care professional at the research or evaluation stage, the For Providers page covers operational and regulatory considerations in detail, including CQC documentation, consent processes, and what to ask providers. Our guide for families may also be useful context for conversations with residents and family members.
For a tailored enquiry, write to hello@humanoidrobotcare.co.uk with a brief description of your setting and what you are trying to achieve.