Would you let a robot into your home? The remote-operator question
The first home humanoids are smart, but not smart enough. When they get stuck, a person somewhere else may take over and see through the robot's eyes. Here's what that means, and what to ask before you ever buy one.
By Duangduean Studios
Imagine a robot that folds your laundry, tidies the kids' toys and loads the dishwasher. Now imagine that, some of the time, it isn't really the robot doing it. It's a person in another city, wearing a VR headset, looking through the robot's cameras into your living room.
That isn't science fiction. It's how some of the first home robots actually work today.
How it works
1X's NEO is the best-known example. It's a humanoid built for homes, priced at $20,000 or $499 a month. Its AI can handle some tasks by itself. For tasks it hasn't learned yet, a trained human operator can take control remotely, using a VR headset to see what the robot sees and guide its hands.
1X has been open about this. The company says owners decide when an operator can take over and what they're allowed to do, and the robot's glowing "ear rings" change colour so you can tell when a human is in control. The remote sessions also help train the AI, so over time it needs less help.
From an engineering point of view, this makes complete sense. Robots learn from examples, and the best examples come from real homes. It's a bit like a new employee working with a supervisor looking over their shoulder.
From a privacy point of view, it's a much bigger deal.
Why this matters more than a smart speaker
We already let a lot of technology into our homes: phones, smart speakers, doorbell cameras. A home robot is different in three ways:
- It moves. A doorbell camera sees the doorstep. A robot can walk into the bedroom.
- It touches things. It can open drawers, pick up documents and move objects.
- A human may be behind it. Not just an algorithm processing data, but a person seeing your home in real time.
None of this means home robots are bad. It means you should ask harder questions than you would about a new phone.
What to ask before you buy any home robot
When home robots reach more countries, and they will, here's a checklist worth keeping:
- Can I see when a human is in control? There should be a clear light or sound, every time.
- Do I approve each takeover? Or is it automatic?
- Can I set no-go rooms? Bedrooms and bathrooms should be off limits if you want them to be.
- Who are the operators? Where are they based, how are they checked, and are sessions recorded?
- What happens to the video? How long is it stored, who can watch it, and is it used to train the AI?
- Can I use it with operators switched off completely? Even if that means it can do less.
- What if the company shuts down? Does the robot still work, and what happens to your data?
If a company can't answer these clearly, that tells you something.
The bigger picture
For a while, many "AI" products will quietly have humans behind them, filling the gaps while the technology catches up. That's not necessarily dishonest. Self-driving car projects worked the same way, with safety drivers. What matters is that it's clear, controlled by the owner, and temporary.
As builders of software ourselves, we think the companies that win the home will be the ones that treat privacy as a feature, not a footnote. A robot you can't trust is just an expensive camera with legs.
Further reading: The first consumer humanoid robot is here, but strangers see into your home (TechSpot) · In-home helper robot NEO uses remote VR operators to help with chores (Road to VR) · 1X gives its NEO home robot faster hands (Runtime Wire)