Ashlee Vance on humanoid robotics: China dominates actuators, US software lead may not be enough

purpose-built tool for planning and building products. We have Ashley Vance in the Reream waiting room. Let's bring in Ashley Vance into the Reream waiting room. It's been far too long. How are you doing? Good to see you. Welcome to the show.

It's so great to have you back.

It's so good to have you back. Uh congratulations on all the progress. What a year. I was laughing about uh that video that we did before we had guests um announcing core memory and putting the the traditional media on on notice. Uh it's been really fun uh watching you grow everything that you're doing. Maybe uh it'd be great to just like reset on the shape of the business right now, some of the stories you've been interested in covering that you've covered recently. And then I I just want to take your temperature on what you're seeing and and the the types of entrepreneurs that you're interacting with.

Yeah. Uh yeah. Well, I don't know [clears throat] which bucket to start with. I mean, we've been running around the country. Um filming a bunch of new video episodes. So, we just put up a bunch of Tennessee, went hard tech, we did Detroit, New England. I just got back from Texas. Those all be coming out. So, yeah. You know me, man. I've been running around chasing uh a lot of hard tech stuff, biotech, all the weird all the weird wonderful stuff. And then I don't know, I got really deep into uh robots and gene editing. I think

that's right. I saw your post about maybe comparing American uh humanoid robotics companies to the Chinese humanoid robotics companies. Uh what stuck out to you as like the important questions to ask? And then uh I' I'd love to kind of tussle with Are you buying would you rather own figure at 39 billion or unree at 7?

I mean, you know, I think I'm I'm going Uni Tree, man. The the you know, this all started I was I was kind of a lark. I started digging into these robot fights in San Francisco and then I think I was I was like shocked that the only robots they could get to do these fights all come from China. And then I started digging into like the parts that go into these. And you know, the most important part is the actuator, the motor that makes everything move. And they're all made in China. I think

I think Tesla made it like a $700 million order for actuators, which was notable for me because I assume that means that Elon's planning to sell a lot of these on like a relatively nearterm time horizon. I don't know.

Yeah. But yeah, I mean, you know, like Tesla sort of has the [sighs and gasps]

Well, I was texting Elon about this last week because I wanted to get to the to the bottom of who [laughter] actually who actually makes actuators in the US. I mean, Elon said sometimes they prototype actuators in China, but they're going to build them in the US. And then, you know, for everybody else, this is a crazy point of weakness, I think, because China is clearly the actuator motor capital of the world and and everybody else is buying them um from them. And so, I don't know, you know, as I dug into this story, I got um I'm not I'm not like, you know, I'm I I enjoy being an American. I'm pretty pro- US. I'm not crazy nationalist, but I I was I started to uh I started to get pretty afraid for the US robotic scene. Do you think we'll see uh any type of regulation around uh Chinese uh humanoids?

I've been thinking about this a lot. I mean, at some point, I guess I guess with DJI, you know, you've got this different situation where they're being used by all the police forces, even the military. I think it's like a much easier case for someone like Skyo or, you know, politicians to come in and say, um, this doesn't make a lot of sense. clearly like at this point of robotics it seems a little less um of a threat to national security but the second the armed forces or anyone's doing serious stuff with them you know I would think unitry would be up next but there's there's like 12 unites as well you know that's the that's the amazing thing that's going on

yeah Brett Adcock was beefing with one of them there was

UB tech yeah

they were beefing back

and they were beefing back saying it was it was real did Did you see that? Missed opportunity for you CGI or did you think it was real?

I didn't I didn't see that video. I've seen I've seen Brett beef me [laughter] with everybody.

Missed opportunity for UB Tech to have one of their robots do like a rap disc on

For sure. Yeah.

Brett and Figure.

Yeah. It was

Yeah.

Yeah. Yeah. Yeah. Sorry. No, no, no. Go ahead. I mean, I I do think it's funny all the like the what do you I mean, I'm curious what other I I'm obsessed with the fighting robots now and I I realize it's like early days with these, but I actually think this is like the most interesting thing happening.

I want the I've been pushing for the the robot like a X games like in challenge like I want to see robots skydiving like uh that's not an X games thing, but broad broad set of

super hard cuz you got to be water resistant too.

Wings. Yeah. big wave surfing wave

and then you also have to swim and you're a heavy heavy robot who might just sink to the bottom of the ocean if you fall off the surfboard. I think that might be the last one.

You could do this versus like the enhanced games and uh and see who wins.

Yeah. G give us your we we sent a couple folks on our team to a local uh humanoid robotic fighting league, underground fighting league. Give us your review. Uh is it is it ready for prime time as a consumer? Uh

yeah, to me to me right now it's like an amazing idea and yet the actual experience like from an entertainment standpoint is probably like a one out of 10 whereas the idea is like a 10 out of 10.

Yeah.

Yeah. I mean it's kind of you know it's it's like a curiosity I think at this point. I mean the the motors are the problem cuz they they all overheat when you throw too many punches

and then the robot the robot stalls out.

What about laundry though?

Better not. [laughter]

This is the thing though. So, like on all these repetitive tasks, they can they can sort of regulate the movement. It it's when you're trying to throw these rapid punches and you're under attack. Yeah. And then the whole robot just freezes up. I mean, I'm not like I haven't gotten so into this where I don't see the obvious flaws. Like, I don't think it's ready for prime time yet because these things just don't don't last that long. But

what about it? That that feels bullish to me because uh if you watch a F1 race like the temperature of the tires matters, the like the wear on the tires matters. And so you're watching not just the pilot of the F1 car, but also the the consumables, right? And the motors are somewhat consumables,

right?

Yeah. It's like, okay, the unit tree is really whailing on whailing on the the figure, but it's overheating.

It's overheating, so it might come back. Is it a one is it a one motor stop or two motor stop type? I mean, I was at one where the robot's leg fell off in the middle of the fight. So, yeah, you could just have somebody come out. How quickly could you get a limb back on?

Okay, so I think

you you have a serious question. You should tell operation.

Well, yeah. So, but one a s a potentially a uh uh product line for core memory is humanoid bench where you get as these things start being available for production you get them up on stage and they do various tasks you know like fruit cutting a fruit with you're throwing a piece of fruit at it watching them you know cut it and dancing and fighting I I I think there's something here but I

actually this is genius yeah it

let's do it uh but but a but a yeah more serious question on teleoperation. From everything that you've seen so far, do you think humanoids are ready to have one in your home that could be remotely operated by someone and and and create any type of value besides novelty?

I mean, like, could it Yeah, like you could do it today. I I find I'm just frustrated by all this. I've been covering teley op stuff for like at least 10 years and um most of it seems pretty similar to what I was writing, you know, videoing and writing about 10 years ago almost. And so um I mean I saw the 1x demos. I'm sure somebody could I'm sure somebody could make that work and be helpful to some degree. I think it you know it probably suffers from all the same stuff as the fights. It kind of falls over pretty quickly. But but you could do something useful. I I it's hard for me like who who Yeah. Like this stuff needs to get better faster [laughter] so that we're not we're not doing that and there's just a robot.

What's going on with Boston dynamics? What's hap what's what's the dynamic in Boston?

Yeah, we got to get you out there to help us understand this.

Yeah, I've never I mean they I've never really dug in on them just because they seem so frustrated that they put out what seems like all the coolest stuff and don't seem to sell much of anything except

a few things to the military. I do not think Boston Dynamics will be the American hope against unitry.

I wonder. Yeah, you'd think that they would at least be set up on some like I I know the company's changed hands a few times. It feels like if you're trying to just, you know, catch up to Unitry, just bootstrapping on top of an existing, you know, it's like it's like what we're seeing today with Gemini 3. Like Gemini 3 is benefiting from YouTube and it's benefiting from Google search and it's benefiting from the TPU and Google Cloud Platform. Usually it's easier to build the new cool thing inside of the organization that has a bunch of resources. Uh but maybe it's a different entirely different architecture or something like that. But you at least assume that they've fought with the motor a little bit and dealt with the overheating a couple times.

Yeah. I mean I was I was with a bunch of robot nerds last week. They were they they were contending. I don't really know where Boston Dynamics is with um with with humanoids, but they, you know, these these robot guys were telling me that dogs are just so much easier than humans because the second the human start walking, you put all this force on the one foot and it's it's like creating all this uh throwing the balance out of whack, putting all this pressure on the motor and that's why um it's kind of easier to pull off some of the parlor tricks.

Interesting. Okay. What's the most underhyped hard tech company right now?

Most underhyped hard tech company. God, that's hard, man. I mean, I'm always I'm always curious to see what Casey Handmer actually cooks up. Um,

he's so smart. I I kind of like believe in the hustle. I feel like the promise of what he's trying to deliver is so massive. That's where my skepticism comes in. But, you know, like so if if Casey um you know, if anyone's going to do it, I I sort of believe in him, I think.

Yeah. I He's somebody I want to win so badly. I want him to win so badly. And it and it does feel like at least let I mean, there's so many people that have a billion dollars. Give him a billion dollars. Let him let the man buy some solar panels and figure out the rest later.

Yeah, absolutely. And then I mean I don't know, this doesn't count as I mean it's hard tech. It's not hardware, but I do think um

New Limit, which is a longevity company, you know, backed by Brian Armstrong and run by Jacob Kimmel, it just

everything I hear about them, I mean, they've just done an incredible amount of science with very few people. And um I think Jacob's got got some surprises coming in the new year.

Very nice. Yeah, we talked to Jacob uh when they did some some sort of launch and uh we were very impressed. He was he was a really great uh great great educator. Really really smart the uh the like what he's working on very very effectively.

What's your favorite data center?

My favorite well I went to Stargate. That was pretty cool. Although uh yeah um I mean Stargate just in terms of like the excitement and the size around it and being

it occurred to me that between John Carmarmac and Elon and Stargate that oddly I think super intelligence is going to light up in Texas but like in a really remote part of Texas, you know. [laughter] And I I found this so I grew up I grew up in Midland, Texas which isn't far from Abalene. It's like

you're a Midland guy.

Crazy.

There's tumble weeds and and all that

Texan intelligence here.

Yeah. I mean,

it's like cracking me up. I'm driving through all these

for hours through all this empty space and then I can just see it, man. One of these data centers, that's where it's going to happen. It's going to be right by some like old oil well. And uh yeah, I find it all kind of comical.

Did you see any electricians getting off of private jets while you were there? [laughter] They had they had a I saw I got off a private jet.

There [laughter] we go.

Not mine. Not mine, sadly. Uh but

not yours yet.

No, but but I saw there were many many many electricians. I just didn't see how they were getting there.

Yeah.

Uh what's going on with EV toll companies? We There's uh I'm curious timeline.

Oh, the Tesla Roadster. Oh, [laughter] I mean, well, on the EV tall stuff, same thing. I feel like I've covered that forever, you know. I went out I think I did the first flight ever with Joby and and uh you you flew in it?

I No, I got to like I went out to their I mean, they literally wouldn't tell me where their secret test site was. And we were, you know, was like kind of close your eyes. We're going to land in this this spot in a helicopter. And I we got to see it. Was it really close your eyes or did you have

How many times have you been blackged, Ashley? [laughter]

I remember they were they didn't want to tell me where the site was.

This is this is a tip for founders. If you want to really impress upon whoever is writing a profile on you [laughter] that what you're doing is really important, you got to be like we can't even show you. And then it's like really like we're at an office park in in Menllo Park.

I did I just went to Helion.

Oh yeah.

And we're gonna have a video coming on them and it was It was awesome. But I got So I got to see their new reactor, but they wouldn't let us shoot it with the camera. And I have to tell you, like that thing was one of the most impressive

pieces of hardware, the roomsized bits of hardware I've ever seen. I'm like, why wouldn't you guys, you know, want to show this? Um,

you you know what? Uh yeah, you not that you need to take requests from me, but uh I want I want some video, some documentary, some footage of those natural gas turbines that are in such high demand right now. They're they're bigger than jet engines. There's these scaled up jet engines. Uh there's this massive backlog. There's three companies and the stocks are, you know, doing crazy stuff. Uh I I want to see inside one of those the the the natural gas infrastructure that's going to go into the data center buildout. I feel like that's something that I'm just waiting. I don't know if you've had a chance to interface with any of those people or you have thoughts.

Not yet, but yeah, when I went to Stargate, I mean, it is crazy, right? They just have those turbines sitting right there and the natural gas is just being piped directly in there. I did some turbines up in uh up by the Arctic Circle was in Sweden one time. They are cool. I don't know that. Yeah. Anyway, it's a good idea. I think

I would just wonder about the bottleneck specifically like everyone's saying like this is going to be the next major bottleneck like we have enough chips we have enough data we have enough algorithms or whatever but uh we have enough land uh but we might not have enough turbines to generate

turbines I mean that was the weird thing about that experience though is like you're you're in you know really old American oil and gas country like it feels so so yestery year and it's just being piped directly in into the future. Um

what what's sentiment like in places like Midland around the data center boom?

I think everyone's like excited to get jobs, you know, and then I think if anyone is prepared for the boom bust nature of where we're probably going with AI, I think these people are because they've lived through it for decades. And and so, you know, it's the same thing out there. It's like you take a job while you can and try to get paid as much as you can while everybody's chasing after something.

Yeah. Do you think that the uh a lot of the headline numbers on the job creation stuff on this on these data centers like ridiculously low? It'll be like, "Yeah, we're spending $50 billion and we're going to create like 25 jobs. [laughter] Sometimes it's like 500 jobs." But does it feel like a little bit different out there because maybe they're not counting like secondary economic impacts of like the guy who runs the gas station is has more business and hire some more people.

Yeah. Well, definitely during the building phase, you're talking about thousands and thousands of jobs just when it's finished. I mean, it is always nuts. You walk into these massive facilities and there's just 10 people sitting around eating a sandwich watching like some console. Uh but but you know I I for somewhere like West Texas um or any you know all throughout Texas it has to be a net gain just because they're otherwise so dependent on the whims of just the oil and gas industry and you've got this whole whole new industry coming in. Um and then definitely they're they're flying people in and out of there all the time to see it. What are your do you ever chat with uh retail investors that enjoy uh deep tech uh companies? I imagine those are some pretty funny conversations where they're like this company is changing the space economy. Like I've actually visited [clears throat] them and they have one warehouse and three people there.

Retail investors should not be allowed to invest in space ever [laughter] under any circumstance. I I am constantly harassed on X by all the A um fans who who are like begging they're in Midland too. They're begging me to go out there. I mean that thing is like a full-on full-on cult um that they have going on. So yeah, I always felt when the rocket companies, [sighs]

obviously it used to be governments that did this and then SpaceX has managed to stay private for a long time in Blue Origin, I think rockets are best developed in private because the second they blow up on the pad, all the retail investors freak out even though it's it's like vaguely a normal course of business and and so yeah, retail in space is is bad bad thing. But I get all these uh I get all these nice notes for people who bought Rocket Lab and Plant Labs early because of my my book or movie.

That's cool. Uh have uh autonomous vehicles tracked how you imagined when you sort you know were covering you know these types of companies and products like a decade ago or is anything

some ways yeah some ways no. I mean, I went to the very first DARPA Grand Challenge and, you know, that was a disaster. The cars didn't go anywhere. Um,

I remember,

say, say more. Who who was actually

competing?

It was crazy, man. You know, so for people who don't know, DARPA, you know, put up this contest, put up a bunch of money to see what we could do with autonomous vehicles. And the biggest teams were university teams like Carnegie Melon was a standout, MIT. Um, but in the very first event, well, I remember Anthony Leandowski was there as like a maybe like a 22year-old and he had

he had a everybody else was doing massive trucks with like a little mini data center in the back and he had a motorcycle. Um, and then in the first race, I can't remember how far it was, but hardly anybody went anywhere. Um, you know, like I think two or three teams went like a few miles. And then and then they redid the race and everyone did way better and some people completed like I think it was it was like on the order of like a 100 miles. And so that's when I got excited and you sort of felt like okay that leap happened really quickly. And then I remember I couple years later I'm hanging out with George Hots and he built his own self-driving car in his garage in like a month and I was driving on the freeway with him and it was working and and yeah, so you know, you have these little tastes and you think it's all going to work. Um I think it makes a ton of sense that actually getting it on the roads um took this long cuz it's it's so hard to do. Although everyone says this so it's not original like we all take this for granted so quickly. It is. It is sort of like amazing to me how well they're working in Austin, in San Francisco, um where I've been. They're just everywhere, you know.

Yeah. What I'm what I'm trying to predict is like what what is the thing that people are hyping now that doesn't work at all that will be totally like a real thing in 10 years, right? And like maybe it's humanoids. Right now it's like hard to take humanoids seriously. But then you think about okay a true 10 years from today. Maybe they are just doing any task that you could want them to do around the house or any task that you could want them to do in a retail setting or factory setting etc.

Humanoids is easily that's the thing I like battle with in my head all the time because it feels like sort of like we talked about before. It actually feels like we've made almost no progress. I see everybody folding laundry and opening and closing microwaves still and it like boggles my mind. And then you look at like the the amount of money that is being invested in this like either either everyone is completely insane or we are about to make massive progress. You can tell in China they're making massive progress on balancing [clears throat] on the movements all those types of things. It's still clearly like the dexterity. And then I think the I think China will eventually probably catch the US in software, but I think they're still so much worse at software than the US is that it's it's kind of like it's holding the field back. So if somebody can can figure that out.

Last question for me. Uh we we've really struggled to uh cover quantum stuff. I mean it's been like up and down, but it feels like [laughter] like Yeah. How do you even go about it? Ashley Ashley could have like an anon that was like the Hindenburg for heart attack [laughter] and you could just

Oh yeah, maybe that would be good.

I don't think it's on brand.

You know what I mean? Because like yes like I like I can't build a humanoid robot but I can go to a

You can build a quantum computer.

No no [laughter] I can't build either but I can look at a humanoid robot and be like okay yeah I would buy that but I I can't do the same thing with the quantum computer. And so it's much harder to evaluate, right? It's like even if it's working, it's like how do I even know if it's working? It's it could just be a normal computer like and just be spitting out normal data,

dude. Like even people in the field with PhDs, they like nobody knows if it's working still. I mean, it's like it's like not a good sign. Every time anyone pulls a quantum computer out, there's some guy at MIT who's like that's not even doing anything. [laughter]

I don't know. Quantum is it's

I'm deeply deeply scarred. I I think I wrote my first story on D-Wave like I don't know like 15 years ago and they were telling me that was that was going to pop out you know be doing

general purpose quantum computing in a couple years. So I'm I'm uh deeply deeply skeptical.

And you know and you know the lesson the lesson is like you should have invested cuz $8 billion company now [laughter] 15 years ago it was probably worth like 20 million and so you could have got in really early but it uh I mean the stock chart looks like this right now. Uh and it's just like yeah you're only you're only one pump away from generational [laughter] wealth.

Well there there's I don't think that they've delivered. There's a tinfoil hat conspiracy around uh some group, you know, figuring out something with quantum which is leading to all these old wallets in crypto like waking up and selling, you know, that never

uh who who knows. Anyway,

uh random final question.

How much would you have to be paid to not use LLMs?

Wow, man. Uh,

forever or like

No, just just while we're paying you monthly. Monthly.

Monthly. Oh, to be paid monthly. Not to use LLMs. Ah, I'd probably do it for like I'd probably do it for like 10K, man.

Damn, that's so that's so bearish. That's so [laughter] bearish for super intelligence. No, I figure I figure I mean I figure I figure that because because for I don't know 10 10 grand you can hire an amazing researcher. One of the most valuable the most if you're building a media company or you're

you're uh you know in in the role that you are the probably the most value you can get out of AI in its current state is research and so anyways that tracks

super helpful but I would take cash. Yeah. [laughter]

Okay. So, so, so any any AI like any any AI doomers out there, if you want a a new marketing channel, you can pay Ashley Vance $10,000 a month. He won't use AI and he'll talk about how [laughter] I don't think you can be bought. I don't think you can be bought. But also, Ashley, have you tried Gemini 3 to the fullest extent?

I have not yet. I'm always

Could change everything. Could change.

Always going back and forth. Yeah,

we would encourage you to. Is Gemini are they they're a sponsor? I think they're coming.

They're coming out as a sponsor for us, too. So, I'm all in. I'm all in. We're going Gemini 3. I'm changing my mind.

Let's [laughter] do it. Also, Sergey Brin was flying his $150 million blimp around San Francisco on the day Gemini 3 beats nearly every model benchmark. You've made a video about this big exact blimp.

I've been pitching Logan at Gemini to to uh make it the Gemini blimp. I They really should. They really should.

Guys, guys, it's not a blimp. It is an airship.

What's the difference?

All right. All right. There's a whole Monty Python video about this. And the the an airship has rigid structure. A blimp is is just a bag. And the airship you can you can do a lot more with an airship. So the a blimp's only ever going to have that tiny little bottom.

Yeah. Yeah. Whereas an airship, you know, you can carry

tens of thousands of tons of cargo with this rigid rigid structure. So yeah, and if anyone ever wants to fly one, you can do it in Germany. Zeppelin still uh flies out by Lake Constants just outside of of Munich. I've done it. It's amazing. I recommend it.

This is amazing. Yeah, people are correcting it on the timeline saying it's not

Dude, you get this is this is like

owned if you say if you call it blimp.

It's bad in aviation. Airship airship. [laughter] I like I like an airship. I'm excited for it. I do wish it had a livery a Gemini livery to celebrate Gemini 3. Uh well,

uh any there's that startup airship industries. Any is that a category that will see a lot of investment, do you think, or or do you think

I mean I've been meaning to meet up with those guys. I mean the airship is like always kind of coming back. It is crazy. Like so be like leading up to World War II,

getting into World War II, I mean there were airships everywhere and you know they were making massive flights from Germany to Brazil. They were carrying thousands of pounds of cargo. I there is a they're just extremely expensive and very hard to make and but there is a whole movement that you can carry tons of stuff and so so less less kind of tourism and more just carrying cargo um kind of like faster than a train but slower than a plane and and they're pretty green.

You need an airship, Ashley. You need you need a studio and an airship that you can just float [laughter] around the US

meeting all these hard tech. You don't need to you don't need a private jet, you know, you don't need to go that fast, but if you could just kind of float between hubs.

I was told that my kids are supposed to be on one of the first flights on Sergeys when it takes passengers. There we go.

So, we'll see.

Uh well,

thank you so much.

We'll join. We'll join too.

Always fun hanging out. Congrats on all the progress.

Yeah, great.

Thank you guys. Congrats to you.

Always a great time.

Thanks, guys.

Have a great rest of your day.

Good to see you.

All right. YouTube.

Up next, we're going back to the timeline.

Eightsleep.com. Exceptional sleep