Digitized | Spotlight
00:05:32 · Michael Cho
[00:05:32] S2: Yeah. So like I say, I think a big thesis of mine even like, I would say nearly four years ago when we decided to really do it full time was data is gonna... Like real world data could be a big big part of of solving Embody AI. And so you asked about, like, how we got into crypto, like, how has crypto involved? And so to be honest, when we... Me and Sam first started, I I thought, like, most skeptical crypto is scam. Like, I was very skeptical about crypto to be very honest. But then I discovered Helium. Like, this is... I remember... Yeah. This was, two months after starting the company. At the time, was only me and Sam. Right? And so yeah, and then Helium just totally blew my mind. I thought it was crazy that you can crowdsource both the hardware cost as well as the actual deployment across the world. There they were selling you know, hundreds of thousands of units, I think, by by that point, you know, at at about $500. And we were also trying to get our robots down to $500. And I wanted that that that hidden map that hidden had, and and so I thought I I need to learn about that. And I would say fast forward to now, I definitely 100% believe in VPN if it's done well. I think some things you can scale a lot faster using, you know, this kind of permissionless incentivization scheme. And I think especially for robotics in... Especially in in the research world, they call this like in the wild data, which is, you know, the the data is collected literally where the robots are supposed to be, right, instead of being in labs and warehouses and and things like that or a a controlled environment. So... And this is very critical because unlike the older days when you do, like, what they call classical robotics, meaning the robots are... These are robots that you'll probably find in factories, you know, like, they are doing the same thing day in, day out, like, you know, thousands of time every, like, every month. So you can use very deterministic kind of, like, coding to get that... Get those work, and those work very, very well. But those don't generalize. Right? But if you use the new paradigm, you know, the most extreme is you train thing what they call end to end. Ultimately, these AI models is just a reflection of the data that you collect from.
00:07:54 · Michael Cho, Winnie
So it makes a big difference if the AI models are trained using data that's collected from various environments. And so in our case, the SAWO robot, we now have robots in more than 50 cities because people are starting to buy them. These are cheap, like like toys. And it turns out that diversity in environments, it's... It it makes a huge difference on building a model that it's already quite generalizable even though you don't have that huge amount of quantitative data, but they are already quite generalizable because of where the data is collected from. And I think crypto is the best way to diversify and and get people around the world to contribute. Yeah. [00:08:36] S1: Yeah. No. I think crypto and kind of like a lot of deepened products have proved one of the early use cases around dramatically reducing upfront capital costs, incentivizing the right players. I'm curious to kinda understand how you guys think about kind of that incentive design. Like, how do people earn? What are the data that you guys are collecting? Walk me through what uniquely suits kind of robotics r and d and how... What the data looks like in comparison to some other projects that maybe listeners are more familiar with, like Helium that you mentioned?
00:09:08 · Amanda
[00:09:08] S0: With BitRobot, we're building a platform for distributed open robotics research and development. We're doing this through what we call a network of subnets. You can think of a subnet as a market that's focused on a particular type of output that's pushing embodied AI forward. So creating a specific type of dataset, facilitating, evaluation of models in the real world. And then the types of data that we're collecting, and across that include teleoperation, video, and simulation. So these are the three types of data that are needed for robotics. And then the benefit of having a network of subnets is there's many different approaches right now to how to move robotics forward. And so we can basically pursue all of them in parallel. So like pursuing teleoperation, but also egocentric video. And then another trend is cross embodiment or co training. And so this is like, okay, let's make sure that the model works on different types of embodiments, different types of tasks. And so we're able to kind of, through our different subnets, showcase, different types of data and then pull that together in something that can be a more general purpose AI model. So yeah, would say those are some of the strengths that we have. I don't know if, Michael, you want to add?
00:10:26 · Michael Cho, Winnie
[00:10:26] S2: Yeah, I mean, to double tap, but it's basically still back to this point I was trying to make earlier, which is diversity. Diversity is not just in the environments where you collect this data, should also be diversity in terms of the robotic types right, humanoids or saurobots or drones. It turns out even if you are gathering data from these very different robotic types, actually you can do a more generalized model that can potentially work across all hybrid robots. [00:10:54] S1: Yeah. [00:10:57] S2: Yeah. And BitRobot is set up in a way that allows that flexibility and and, know [00:11:02] S1: Very interesting. Today, I think I read that anyone can kinda contribute to subnet one, which is, like, basically a Pokemon style go sidewalk where we gather real world human... Real world urban data and then subnet five. What has been the response so far? And maybe you can tell folks a little bit more about the two of those.
00:11:02 · Winnie, Michael Cho
[00:11:02] S1: Very interesting. Today, I think I read that anyone can kinda contribute to subnet one, which is, like, basically a Pokemon style go sidewalk where we gather real world human... Real world urban data and then subnet five. What has been the response so far? And maybe you can tell folks a little bit more about the two of those. [00:11:22] S2: Yeah. I can talk about the first one. So because I've been doing the longest and cyber robot. So it is actually quite a fun game. It just... Yeah. Like you say, it's it's very much like Pokemon GO in the real world. So the idea is, like, the cyber robot is literally like an RC car with four gs basically. So what that means is that you can be, let's say, sitting in your home while controlling a robot either around your neighborhood or even halfway across the world because these things are just running, you know, through the Internet. And that makes it super fun because, you know, it it happens in the real world. You're collecting some virtual items, you know, on the back. But, yeah, this happens in the real world. So I think it makes... Yeah. I I find it, like, I still find it really fun to to drive. You know, I don't drive it as often as I when I first started out. I've been to many cities that I would probably would never get a chance to go to. I remember the most memorable one I tried was, it was super hot one day in Singapore and then I was driving at this robot that we have in Sweden. Was snow everywhere, magical, it was so magical, but it's probably a very ordinary place like [00:12:33] S1: you
00:12:33 · Winnie, Michael Cho
[00:12:33] S1: you [00:12:33] S2: know the suites are probably just used to snow everywhere. I was like sweating in Singapore but it was. So that... Yeah. I think that aspect is And [00:12:44] S1: And how do people access it? Is it just kind of a platform, a portal, app, website? Walk us through what that looks like. [00:12:53] S2: Yeah, from the gamer perspective is right, just think of it like a cloud based game. Like you assess it on the browser, you don't need to download any app, just log in. Yep. And then just drive. So what you see is the front camera, the back camera. [00:13:09] S1: Yeah. That's [00:13:09] S2: awesome. Link it up with your game controller, you start driving. Yeah. [00:13:12] S1: And, like, how has the response been so far from, you know, the community and folks who have been kind of playing around with it and kind of helping you guys contribute and gather data?