Rivian Automotive, Inc. Class A Common Stock Evercore's 9th Annual ADAS, AV & AI Forum

Yeah. Well, first, thank you so much for hosting us today, and great to be here to be able to talk to investors. So, in terms of the key areas, I can talk a little bit, and then James, we'll pass it over to you on autonomy. But really what our investors are focused in on right now first is. Sorry, I'm getting some feedback. All about the ramp of the R2 right now. So that's our midsize SUV that we launched back in June. As you can imagine, this Q3 being the first quarter that we'll have really had a full quarter of production. Lots of questions that I have no doubt we'll talk about of, how is the ramp going from a production and a supply chain perspective, the targets that we laid out on the Q2 earnings call.

Then of course, a lot of focus on autonomy and our path there.

Yeah. The team's been laser-focused on getting point-to-point ready this year. So, we're very much on track there for initial rollout towards the end of this year. We've really been on this autonomy journey since the launch of Gen 2. Gen 2 was built around this idea of vertical integration and a very strong data loop so that we can collect customer data, customer sensor data together with the driving action. That's really what's allowed us to drive the autonomy development. We launched Universal Hands-Free last year, end of last year. That's the nation's largest scale hands-free system that you can get. With point-to-point, we really think this will be the next important step in our autonomy journey, yeah, as we iterate towards those kind of higher levels of autonomy in the stack.

Yeah. We believe very strongly in multimodality. We think that even if you accept the premise that cameras could be comparable to a human-level perception, we think there's many cases where, actually, that's not enough, where humans also make mistakes. We think that the societal bar will almost certainly be better than human level for widespread autonomy. For those reasons, we think that, yeah, multimodality really helps in the field. It also helps us accelerate our progress. What we see is that when we have radars with cameras, it actually makes the system that much better. When we add LIDAR, it makes it better again.

This is really a way to accelerate our progress, as we're building the scale with R2, as well as providing a set of sensors that really provides a safe and performant and robust capability to our customers.

Yeah. We started, let's see, three years ago, where we brought the full autonomy stack in-house. Everything from the lowest level camera drivers and ISPs, all the way to the perception models running on the vehicle. Gen 1 had a Mobileye system. With Gen 2, when we launched that, we pulled that system out and replaced it. Then, as we've been gathering that data, developing our perception models, we've been steadily improving things. First of all, with our Highway Assist feature, then Universal Hands-Free, and now the big focus is on point-to-point.

Yeah. Autonomy is really about kind of the edge cases. A big focus for us has been not how can we get more boring highway miles, which are not very informative for the model. The sort of gradient you get back when you train with that data is very minimal. But how we can find more interesting cases, which are really sort of driving, kind of different behaviors from the model. The gradient is large. As part of that, we developed this grids system where we can, actually using VLMs, modern AI techniques, we can scale up the collection of edge cases. We currently have a 3,000-query set of very long-tail edge cases that are constantly querying against the customer fleet, and then we're using that to train the model.

I think that's a big part of the development process, is actually getting those edge cases, getting new things that the model hasn't seen before, so that it can refine its driving behavior.

Yes. We will begin the rollout of point-to-point at the end of the year to probably our early adopter customers, and then we will continue that rollout into next year, with the aim of having it rolled out to the full Autonomy+ customer fleet a little later next year. As obviously we are rolling that out, we are continually developing and improving the system. We will get improved disengagement rates, more scenarios with better handling. There are also some features we probably will have out of ODD for the first version that we will aim to add in, for example, parking lots. A big focus then is on the Uber Robotaxi partnership, where we are then iterating on that system, getting it to a fully driverless level of performance. That is both on the model side, but also on the validation side.