Are People Comfortable With Home Robots?

That question usually gets answered with vibes: cute demo videos, scary robot clips, or surveys about hypothetical humanoids. RAI Institute's public Spot pop-up is more useful because it gave thousands of ordinary mall visitors a short, direct robot interaction and then measured how their attitudes changed. The result is not a blank check for home robots. It is a reminder that hands-on experience can make people more realistic, not just more excited.

What did the Spot test-drive study actually show?

In summer 2025, RAI Institute set up a free robot experience at the CambridgeSide mall. The pop-up included a museum area and an interactive Drive-a-Spot arena where visitors could control a Boston Dynamics Spot quadruped with a simplified adaptive game controller. The controller had large buttons for basic movement, height, sit, stand, and tilt commands. IEEE Spectrum's report said the people who drove Spot ranged from age 2 to over 90, while the surveyed participant group spanned children through older adults.

The useful part for home-robot buyers is the before-and-after measurement. Roughly 10,000 guests visited the Robot Lab. About 10 percent drove Spot and opted into the survey. More than 65 percent of surveyed participants had seen Spot in images or videos before, but most had never encountered one in person.

The survey asked how comfortable people would feel encountering a robot in five contexts: factory, home, hospital, office, and outdoor or disaster response. It also asked how suitable Spot seemed for each setting.

The pattern was clear: after a short hands-on driving session, comfort scores increased significantly across all five contexts. Home comfort started roughly neutral, then rose significantly after people controlled the robot. Perceived suitability also increased, with the largest gains in the more socially delicate settings: home, office, and hospital.

That does not mean everyone suddenly wanted a robot dog in the living room. The effects were described as small to moderate. It also does not prove the attitude shift lasts for months. But it is still a rare piece of evidence in a market full of polished videos: direct agency changed how people judged a robot.

Why videos are a weak test for robot trust

Robot videos are excellent at creating emotion and poor at setting expectations. A tightly edited humanoid demo may hide resets, remote assistance, controlled lighting, or a room arranged specifically for the task. A viral robot-dog clip may make a quadruped look threatening even when the actual product is an inspection machine, a research platform, or an awkward but supervised demo.

The RAI result matters because most people had already seen Spot online. Video exposure did not give them the same calibration as a few minutes of control. Once people felt the robot respond to simple commands, crouch under barriers, step over obstacles, and adjust its body in a tight arena, they appeared better able to imagine where the robot fit.

That is exactly the buyer problem for home robots. You cannot judge a moving robot the way you judge a smart speaker. A home robot has motion, noise, size, turning radius, camera behavior, recovery behavior, and social presence. It may look smaller in a product render than it feels in a hallway. It may look scary in a security clip but feel less intimidating when it obeys a slow, obvious control scheme. It may seem charming in a launch video but annoying after the third time it blocks a doorway.

The lesson is not "try every robot before buying" because most people cannot. The lesson is to demand evidence that approximates direct experience: uncut task videos, live demos, public trials, return windows, rental options, clear noise and size specs, and honest failure behavior.

The home is the most sensitive robot environment

Factories and disaster zones are easier places to justify robots. The tasks are clear, the spaces can be controlled, and people already expect machinery. Homes are different. A robot in the home is not only a tool; it is a guest with wheels, legs, sensors, microphones, cameras, and software rules.

That is why the home-context gain in the Spot study is interesting. People did not merely decide Spot was more suitable for factories after driving a factory arena. They became more comfortable with robots in the settings where baseline skepticism was higher. RAI's researchers suggest that hands-on control may shift something broader: a person's understanding of what the robot is capable of and where it might be appropriate.

For a buyer, this means acceptance is not only about friendliness. It is about legibility. A robot feels safer when people can predict what it will do next, see why it stopped, understand its boundaries, and intervene without drama.

That is where current home robots vary sharply.

What the emotional data says about companion robots

The post-interaction results were more emotional than a spec sheet. IEEE reported that 74 percent of participants selected excitement after driving Spot, 50 percent selected happiness, and only 12 percent selected nervousness. More than 55 percent rated the experience as "brilliant," and 62 percent said they were very likely to recommend it to a friend.

Open-ended responses are even more relevant to companion robots. The most common memorable moments included locomotion and terrain adaptation, mentioned by 22 percent, and expressive tilt movements, also 22 percent. A smaller 3 percent described anthropomorphic reactions, such as worrying about hurting the robot or finding its behavior silly in an emotionally meaningful way.

That is the bridge from utility to companionship. People may start by asking whether a robot can carry, inspect, or clean. After direct interaction, they may notice movement, body language, and play.

The study's task-preference shift points in the same direction. Before driving, people's desired robot tasks clustered around domestic assistance and heavy or hazardous work. After driving, domestic help stayed important, but entertainment and play jumped from 7.5 percent to 19.4 percent. Companionship appeared at 5 percent.

This does not mean a quadruped becomes a good elder-care robot after one mall demo. It does explain why products like LOVOT, Sony aibo, and PARO should not be judged only by chores. ui44 tracks PARO as a therapeutic baby seal robot used in care settings, with tactile, light, audio, temperature, posture, and microphone sensing, plus FDA Class II medical-device status in the U.S. Its job is not utility. Its job is acceptable, repeatable interaction.

What buyers should ask before inviting a robot home

The Spot pop-up suggests that people become more comfortable when the robot's behavior is legible. Buyers can turn that into practical questions.

Can I understand the robot's motion before it moves?

A home robot should signal intent. A rolling robot should make it obvious when it is about to leave the dock, enter a room, follow a person, raise a camera, or turn around. A legged robot should move slowly enough around people and pets that its next action is predictable. A humanoid should be even clearer, because arms and hands raise the stakes.

Spot is not a consumer home robot, but its database record shows why motion specs matter: 33.8 kg with battery, about 90 minutes of battery life, 1.6 m/s top speed, IP54 weather resistance, stereo cameras, time-of-flight, ultrasonic, infrared, laser ground sensing, and a 14 kg payload rating. Those numbers shape how it feels in space. The same principle applies to smaller home robots. Size and movement are not secondary specs; they are trust specs.

Can I stop it, steer it, or set boundaries?

Agency was central to the RAI experiment. People controlled Spot instead of just watching it. Home robots need a version of that agency: physical stop controls, app-visible no-go zones, guest mode, camera and microphone toggles, and obvious manual override.

Astro is a good example of why this matters. Its strongest home role is patrol and remote monitoring, but that also makes it a mobile camera platform. The product is easier to accept when users understand Visual ID, Ring integration, room boundaries, and when monitoring requires a subscription.

What does the robot do when it fails?

A robot that stops safely and explains itself is easier to live with than a robot that bluffs. Does it ask for help? Does it save the scene for later? Does it call a remote operator? Does it avoid private rooms by default? If a buyer cannot get plain answers, the robot is not ready for trust-heavy spaces.

This is especially important for humanoids. 1X NEO is one of the most home-focused humanoids in ui44: $20,000 pre-order pricing, a 167 cm / 30 kg body, roughly 4 hours of runtime, cameras, depth sensors, tactile skin, and a soft body design. But a humanoid's acceptance challenge is bigger than its spec sheet. Buyers need to know when it acts alone, when it asks, and what human data it uses to improve.

Why public demos may matter more than another hype video

The next useful home-robot proof may look boring: a mall booth, a library trial, a senior-center pilot, a retailer demo cage, or a short home rental. That kind of exposure lets people discover the parts that videos hide. How loud is it? How close does it pass? Does it feel like a tool, a pet, a toy, a camera, or an intruder? Can a child, older adult, or nervous guest safely understand it?

Manufacturers should want this too. If a few minutes of direct control can raise comfort and perceived suitability, then public trials are not just marketing. They are product research. They reveal what people misunderstand, which motions feel expressive, which contexts feel invasive, and which tasks people imagine after the robot becomes physically real.

For buyers, the RAI result is a filter. Be skeptical of robots that only exist as cinematic clips. Give more credit to products that provide unedited demos, transparent limitations, hands-on access, specific specs, privacy controls, and clear service paths. A robot you can understand is more valuable than a robot you can only admire.

So, are people comfortable with home robots?

Not automatically. The home remains the hardest acceptance test because it mixes utility, privacy, safety, emotion, and intimacy. A robot that is welcome in a warehouse may feel strange in a hallway. A robot that is cute on a desk may be unwanted near a bedroom door. A robot that helps a caregiver may be unacceptable if the person being cared for feels watched.

But the Spot pop-up suggests comfort is not fixed. When people move from passive exposure to direct experience, their judgments can become warmer and more practical. They may still worry about the right things: cameras, control, safety, failure, and purpose. That is a healthier form of acceptance than blind hype.

The best near-term home robots will not win merely by looking friendly. They will win by being understandable: clear motion, clear boundaries, clear fallback, clear privacy, and a job that makes sense in the room where the robot actually lives.