A humanoid robot that folds your laundry and loads the dishwasher sounds like the future arriving on schedule. The first one sold to ordinary households, 1X's Neo, costs $20,000 and is due to ship this year. The catch is that when Neo gets stuck, a person somewhere else may be driving it. That detail matters far more to a business owner than the robot's looks.
What happened
1X opened preorders for Neo, a soft-bodied humanoid built for household chores, and says the first units will ship in 2026. In a post about its new California factory, the company says its plants can build up to 10,000 Neos a year, but that the robots now coming off the line are going first to its own team for in-home testing. It says some customers will get one this year and some later. Read the company's factory update.
The bigger story is how Neo works. MIT Technology Review reported in February that Neo can be piloted remotely by 1X staff when it hits a hard task, that the company gets the owner's consent before switching to remote control, and that its founder is not committed to any set level of autonomy. The same piece describes workers in other companies wearing headsets and sensors to repeat simple tasks hundreds of times so robots can learn from them. MIT Technology Review on the hidden human work behind humanoids.
A follow-up from MIT Technology Review on October 8 was blunter. It reported that a remote human operator currently performs most of Neo's tasks, and quoted Agility Robotics cofounder Jonathan Hurst estimating about ten years before robots do useful work in homes. It also cited figures that roughly 15,000 humanoids shipped in 2025, about 90% from Chinese companies, and that Agility's robots in trials at firms such as GXO and Amazon mostly move bins and totes. MIT Technology Review on why robotics breakthroughs won't change your life soon.
TechCrunch reported in January that 1X released a "world model" meant to help its robots learn from what they see, while a spokesperson declined to share a shipping timeline or order numbers. TechCrunch on 1X's world model.
Why it matters for your business
Remote operation is not a scandal. It is a sensible way to get a hard product into the world while the software catches up. But it changes what you are buying. A teleoperated robot is part machine and part remote worker. That has three consequences.
Privacy. If a stranger can see through the robot's cameras, the robot is a camera on wheels with a human behind it. In a home that is awkward. In a law office, a clinic or a store back room with customer records, it is a compliance question.
Labor, not just hardware. The price of a humanoid includes someone's time, now or later. When the vendor's remote team is the real worker, your service quality depends on their staffing, their hours and their training.
Honest expectations. Demo videos do not say which moves were autonomous. The MIT Technology Review reporting notes that many impressive demos are scripted or remote controlled. Buying on a video is buying on a promise.
The other side deserves a fair hearing. Every new category starts rough, and remote help is how the robot gathers the data it needs to improve. Early buyers are, in effect, paying to train the product. For some that is a fair trade.
What we'd do
- Ask the remote question in writing. Before any robot pilot, get the vendor to state which tasks are autonomous, which may be remote controlled, who the operators are and where they sit.
- Map the cameras. Treat a robot like any other connected camera. Decide which rooms and records it may never see, and confirm the vendor can enforce that.
- Start with boring jobs. The proven work today is moving bins, totes and carts in controlled spaces. That is where a pilot can pay off.
- Price the whole service. Compare the robot plus remote support against a part-time hire or a simpler machine, such as a conveyor or an autonomous cart.
- Wait on the home models. For most owners, the right move on household humanoids in 2026 is to watch the first customer reviews.
Our view: humanoids are coming, but the near-term product is a robot with a remote worker attached. Buy it that way, with clear terms on who is watching and who is doing the work, or wait a year and let early adopters find the rough edges for you.
Want a system like this in your business?
A discovery conversation is where we map it: what to automate first, where the human checkpoints go, and what it would take.
Book a discovery conversation →