00:04:08 · Ulrik Hansen
[00:04:08] S0: So you're taking an example for self driving, for example. I think the... One of the main challenge that they have, right, so if you take Waymo as an example, they are, like, now live in San Francisco, but they haven't yet expanded to cover the geography all the way to San Francisco Airport. So now the question is, like, why is that? Like, why don't you have, like, fully unsupervised self driving given that it works so well and doubts on SF? And it is really, like, down to real environments being dynamic and they change and eliminate, like, all the different types of edge cases that the car might encounter in the real world is extremely difficult. And so you need to, like, basically factor that into the way that you build, like, your old data orchestration pipeline so that when the car, like, encounters, like, a new type of, like, unseen scenario that you have, like, either, like, a heuristic to have the car solve that problem itself or you have some way of, like, relaying the feedback back into the into the model. And I think that's something that Tesla is doing really well. Right? Because they have now supervised self driving and, like, I have a test myself, but I'm driving the car. Like, sometimes I have to intervene, and that data is sent directly back to Tesla. So they build this, like, iterative feedback loop between, like, okay, actually solving the long tail as it happens. Whereas, like, of course, like, these, like, robotaxis and, like, unsupervised types of use cases, it's much more difficult because they're relying on using their own drivers, right, to, like, go and and collect that data. So, yeah, that that would be one that would be one example. Yeah. Go ahead.
00:05:32 · Ksenia Se
[00:05:32] S1: These are two very different strategies. Like, we talk about Waymo and Tesla because Tesla relies basically solely on computer vision and machine learning. So what's your bet? Because I I just published a little video about self driving, and I was recording video while Tesla was driving me around. So, yeah, what's your take? Do you think Tesla's strategy with more data, with more learning, will win or Waymo more, like, you know, step by step careful approach Yeah. From a data perspective.
00:06:02 · Ulrik Hansen
[00:06:02] S0: Yeah. No. I think the one thing that, like, Tesla really has going for them is they just have a huge and compounding data advantage over, basically, like, every single other player in the market. They are collecting live human feedback data. They have, like, the equivalent of, like, a thumbs up and then some... A thumbs down button that you're used to enchantability, right, embedded in the cars because they have people driving the cars. And so, like, this data advantage is, like, compounds massively over time, whereas, like, Waymo, they have to rely on, like, collecting their own data. Right? They don't have, like, the distribution advantage that that the Tesla has. So, inevitably, like, I think that's a much slower rollout and a much slower build. Like, that being said, of course, like, they've sort of decided to add additional sensors on top of, as you pointed out, like, Tesla's relying on RGB, additional sensors to help inform the planning and trajectories of, like, how the car, like, drives and operates, which I think, quote, unquote, like, should be a safer option, but, of course, it's way more expensive and, like, takes longer to build. I am, like, very bullish on on Tesla, actually, like, getting ahead on on this. And now they're starting to, of course, like, roll out their own, like, robotaxi fleet, which they can probably, like, do very quickly. Now the question is they, of course, like, probably don't have the same advantage or, like, the same lead with working with the regulators that that Waymo has. So I think that'll be an interesting dynamic to to observe over the next, like, three to five years. It's like these systems actually, like, go live in, like, the different cities around The US.
00:07:20 · Ksenia Se
[00:07:20] S1: Do robotics have the same bottlenecks as, autonomous systems like self driving or it's different?
00:07:25 · Ulrik Hansen
[00:07:25] S0: Yeah. No. I think, a lot of the... They have a lot of the same constraints. So I think, ultimately, the difficulty in deploying, like, these, like, robotic systems is, like, you have the additional constraints of you have to deploy the model on, like, specific, like, hardware devices on edge. Right? So you have GPU constrained environments that you don't have your... If you're leveraging a foundation model like a chat interface. You have to, first and foremost, like, make sure that, like, your systems can actually work on these, like, embodied AI devices, which is, I think, no small feat of of engineering. And then, like, creating the iterative feedback loop between the actual product usage and the model itself is also much, much harder because you might have, like, a robot folding laundry in your house, and it might fail. So now, look at, okay. How do you relay that back into, like, the system and, like, how it's trained? It becomes, like... There's, of course, like, latency for, like, uploading the data. You might, like, be in, like, an area without, like, Wi Fi, so you don't necessarily have Internet connectivity. Whereas, like, where we have seen the fastest progress in AI has been, like, in the pure digital workflows. Right? So for language models, like, ChatGPT and and otherwise, for coding, where, like, the feedback loop between, like, the actual user and the model itself is instant. That's another luxury that you have in physical and and embodied AI. So the companies that are building those types of domains, like, they have to really think carefully about, like, how they collect that feedback. And we see it, like, being solved in a number of different ways. Like, some companies are adding 10 operations to, like, help steer the robot if they kind of go off the rails. But, ultimately, like, if you want them to work on a fully autonomous basis, right, you don't necessarily, like, will be able to rely on that. So being clever about, like, how you build those, like, feedback loops is gonna be what determines a lot of the future compounding advantage.
00:08:59 · Ksenia Se
[00:08:59] S1: Feedback loops, we will see more and more of that in AI agentic systems. Right? How does infrastructure evolve to address that?