Written by Silvia Pavelli
1X’s NEO home robot is on sale to US households for a $200 deposit, marketed as autonomous, soft and quiet. The company’s own product page also explains what happens when it meets a chore it does not know: you schedule a human expert to guide it remotely.
The consumer humanoid has arrived, and it comes with a caveat printed by its maker. NEO, the home robot from 1X Technologies, is available to US households for a $200 deposit, with reported pricing around $20,000 to own or $499 a month. The pitch is domestic: take on the boring and mundane tasks, take a list of chores and a schedule, and come back to a cleaner home.
The hardware is designed around safety rather than capability. NEO is described as lightweight, quieter than a modern refrigerator, wrapped in a deformable 3D lattice for cushioning, and driven by tendon actuators for gentle movement. Those are the right priorities for a machine sharing a house with people.
The specification that matters most, though, sits further down the same page. NEO works autonomously by default. For any chore it does not know, the user can schedule a 1X Expert to guide it, “helping NEO learn while getting the job done.”
Read that as a product statement rather than a footnote and the category looks different. What a household buys is not an autonomous robot but a robot plus on-demand human assistance, and the human is the component that scales with the diversity of tasks. A robot that handles known chores alone and unfamiliar ones with a remote pilot has an autonomy rate, not an autonomy claim — and the rate depends entirely on how well the world matches its training.
That arrangement is a human-dynamics problem before it is a robotics problem, which is the layer Brainlayer models. Attention, readiness to intervene, the decision to step in and the recovery that follows are precisely the variables that decide whether an unfamiliar task succeeds or fails — and they are the ones most autonomy stacks leave unrecorded. NEO’s Expert Mode makes that layer visible: the robot logs the chore, but the expert holds the judgement.
This is the honest shape of the technology, and it deserves credit for being stated. The alternative is to promise general capability and quietly rely on teleoperation, which has been the pattern in earlier home robotics efforts where operators worked remotely through the machine’s sensors.
It also raises questions the industry has not answered. If unfamiliar chores require a remote expert, the marginal cost of novelty is human labour, which caps the economics of an expanding task list. A remote operator with access to a robot’s cameras and microphones inside a private home is a privacy arrangement as much as a technical one: who watches, what is recorded, and how long it is retained. And “learns over time” is a statement about future capability, not a present one.
The competitive picture sharpens the point. Tesla has said it aims to make Optimus available to outside customers as early as 2027, and Chinese manufacturers are scaling production aggressively. The consumer segment is where autonomy claims are least verifiable, because the environment is uncontrolled and the tasks are open-ended.
Three numbers would settle the argument, and none is currently published: the intervention rate per chore, the minutes of remote assistance a household consumes per month, and whether the quoted price includes that assistance or bills it separately. Those are the measurements Brainlayer argues should define the field, because they are the ones a buyer can verify. Until they exist, the sensible way to read any consumer humanoid is to look for the human in the loop. There usually is one, and finding it tells you more about the product than any demonstration video.
