Bell Lin (3:29)

So there are certainly trade offs to keep in mind. What about the benefits then that you discovered when testing out these different AI browsers? What were the main things that you found useful?

Nicole Nguyen (3:39)

So the most futuristic element to these AI browsers is that some have agenta capabilities. Perplexity's Comet and OpenAI, which just launched their AI browser recently called ChatGPT Atlas, have the ability to click and type around for you using an AI bot. So they take over your tab and you can watch this robot painfully slowly, look up flights or do whatever you task it with. And there's a kind of magic to letting it do a tedious task, like finding all the ingredients for a recipe that I wanted to make and adding it to my whole foods cart on Amazon, for example. But there are also new risks with this kind of magic. A new kind of hacker attack, called a prompt injection attack, hides secret instructions for the AI chatbot which are malicious in the URL or the website itself. And so you, you can really only do this on websites that you trust. But the beauty of using an AI browser versus using an agent inside of an AI chatbot is you don't have to hand over your credentials to the AI company in order for it to do its work. I'm already in a logged in state on Amazon or Google Calendar, and so it can work agentically without me having to compromise my own security.

Bell Lin (4:56)

How quickly do you think these browsers will take to catch on with people more broadly?

Nicole Nguyen (5:01)

It remains to be seen which app in particular will take off. But the incumbents, including Google Chrome, which has the biggest market share in the world, is releasing its own AI chatbot called Gemini in Chrome. It's rolling out to most US users. And so I think that once people start to use AI in their web browser and get used to that in an incumbent, maybe they'll be more willing to switch to Comet From Perplexity or ChatGPT Atlas from OpenAI.

Bell Lin (5:33)

That was WSJ personal tech columnist Nicole Nguyen. Coming up, robot helpers are the stuff of sci fi fantasy and sci fi horror. No matter which side you're on, the fact is they're coming soon. We examine how one humanoid robot on the market actually works. After the break.

Katie Pizzolato (5:59)

Quantum computers are poised to help businesses drive innovation through more expansive and multidimensional computations, says Katie Pizzolato, vice president, IBM Quantum Platform, which sees a roadmap to these use cases.

Unnamed IBM Quantum Expert (6:10)

By 2029, they unlock a new set of mathematics and algorithms to tackle applications outside the reach of classical computers. Working alone, it really is a fundamentally different way to process information.

Katie Pizzolato (6:21)

Different but not impenetrable.

Unnamed IBM Quantum Expert (6:23)

It does feel more complex. For sure it is, but it is much more accessible and the barrier to entry is much lower than people anticipate.

Bell Lin (6:36)

Standing at 5ft and 6 inches tall, the company One X's Neo robot is expected to ship to households next year. But before you rush to buy one, there are plenty of limitations and risks to be aware of and a few things to get excited about. WSJ senior personal technology columnist Joanna Stern is here to help us investigate how the Neo humanoid robot actually works, the subject of her latest column and video, including why there's an actual human involved in its operation right now. Joanna, let's start by talking about the robot itself. What does it look like?

Joanna Stern (7:15)

If you've seen humanoid robots, you have seen the kind that look like they work in an industrial factory, and Neo does not look like that. My first question to the CEO of 1X actually and to Neo was Neo, why are you wearing a sweater? It's 70 degrees out in California. And the answer was really good, that this is for aesthetics. We want to make this a more welcoming, cozy robot, but it's also for safety. And that's one of the big things that's different about Neo than some of the other humanoid robots that are being developed right now is that they really are making this for the home, with the idea that it's not going to just work in factories, though they say that it will work in theirs. But their real focus is to build a robot that can do chores around your house, empty the dishwasher, clean the counter, eventually cook, do your laundry, fold your laundry. And for them to do that, they really emphasize that it needs to be safe because robots can fall down, robots could smack their arm and someone could be standing behind them. So they've tried to build this robot to be as close close to a human as possible, and they're using special types of Gears and tendons to mimic what human muscles and human biology looks like.

Bell Lin (8:22)

And you mentioned things like folding laundry and doing dishes. What are some examples of things that the Neo robot can currently do?

Joanna Stern (8:30)

The Neo robot physically is capable of doing a lot of things. Now. It doesn't necessarily do those things very well yet. And so there's this idea that The 1x CEO Bert Barnich talked about in our interview, which is robotics slop. It's the most useful kind of slop, because if. If you put all of my glasses from my dishwasher in my cabinet, I'm pretty happy. It is going to be not perfect, but back to, like, just incredibly useful. And so this idea is that, hey, like, other AI slop, it's not great, but it's fine. And so, as you also see in the video, it folds my sweater. It's not great at folding the sweater. And it puts the forks and cups in the dishwasher. Not great at that either. It took the robot five minutes to load three things. So, look, the idea is that eventually it's going to do all of these things perfectly, just like Rosie and the Jetsons. But that's not where we are today.

Bell Lin (9:26)

And can you talk a little bit about how the Neo robot works in terms of having a remote operator there in the background, peering into your living room or your kitchen or whatever room it may be?

Joanna Stern (9:37)

So this is where it gets pretty wild and pretty crazy. And one of the reasons I wanted to do this story is that everything I saw Neo do during the day that I spent with it at its headquarters was being operated by a tele operator. And this is really when a human in a VR headset with controllers controls the robot remotely and does pretty much everything that the robot's doing. So if you move your arm, the robot's arm moves. And so everything I saw it do, from clean the counters, to put the dishes in the dishwasher, to water the plants, to fold that shirt, was done by a VR operator or tele operator, a 1x employee. And so that was a lot of my conversation with 1x was, okay, you're shipping this in 2026, but right now, this robot isn't doing anything really autonomously, which is obviously the goal of home robotics, right? We want these robots to do things on their own, not with a human operator, across the country or across the world. And so the CEO told me that right now, many things this robot will be able to do on its own, but there'll be many things it needs help with which will require teleoperation and will require an employee of their company or representative of their company taking over the robot, being able to see through its camera eyes into your house and do things, which is a crazy privacy issue.

Bell Lin (10:56)

Besides giving up your privacy to a great extent, what are some of the other risks or potential downsides to the Neo robot? And what are the guardrails that 1x has put in place to prevent some of those things?

Joanna Stern (11:09)

They have taken both precautions on the hardware end, but also on the software end to make sure that NIO right now doesn't lift a lot of heavy weights or objects. Nio at first won't be able to pick up anything sharp or heavy or hot. Though eventually, as we trust Nio and these guardrails get better, it should do those things right. Like you'd want it to pick up a knife to put in the dishwasher. You'd want it to pick up something very heavy and help you cart it around the house. But for now, they say there's a 50 pound weight limit on what NEO can lift, though the robot's actually capable of lifting far more.

Bell Lin (11:43)

What are some of the benefits of allowing the NEO robot into your home? And what else would an early adopter seek to or benefit from early on?

Joanna Stern (11:52)

Early adopters, you are signing up right now to train this robot. You are not signing up for a super useful robot. And as you look at this thing, it looks a lot like a toddler trying to figure out the world and you to teach your toddler. It takes many years. And so that's what you're signing up for here. You're not signing up for great utility or usefulness. Will that come? Well, that's the bet the company's making. The company's betting that you will give it training data. All that video of NEO working in your house and loading the dishwasher are videos that then end up in Neo's world model, which is how they train it. Large language models required all the text of the Internet, so much data and that's all out there. But these robots don't have that kind of data. And so that's what 1x is trying to do here. So you are really helping beta test and build the training data for the robot eventually to be great in your house.

Bell Lin (12:41)

That was WSJ senior personal tech columnist Joanna Stern. And that's it for Tech News Briefing. Today's show was produced by Julie Chang. Jessica Fenton and Michael Lavelle wrote our theme music. Our supervising producer is Katie Ferguson. Jessica Fenton is our technical manager. Our development producer is Aisha El Moussleem, Chris Sinsley is the deputy editor and Falana Patterson is the Wall Street Journal's head of news Audio logging off for the weekend. I'm your host, Bel Lynn. We'll be back later this morning with TNB Tech Minute. Thanks for listening.

Katie Pizzolato (13:26)

The time to develop strategies around quantum computing is now, says Katie Pizzolato, vice president, IBM QU Quantum platform just like.

Unnamed IBM Quantum Expert (13:34)

AI was experimental in its early days and is now foundational, Quantum is moving through a very similar arc. So building quantum literacy is important. Educate leaders and technical teams on the basics and implications of the technology and run experiments and identify where you define value.

Katie Pizzolato (13:50)

Quantum, she says, should be seen as a new part of your ecosystem, not a substitute for it.

Unnamed IBM Quantum Expert (13:55)

Think about how this technology is going to integrate with AI and classical systems. Quantum is not going to replace those things. It will augment them. We've got to find the parts of the workflow where quantum is Most valuable.

Katie Pizzolato (14:07)

Visit IBM.com to learn how quantum computing will accelerate business innovation and growth.

Joanna Stern (14:12)

Custom Content from WSJ is a unit of the Wall Street Journal Advertising department. The Wall Street Journal news organization was not involved in the creation of.