AI, R2 and the Future of Everyday Driving | Rivian CEO RJ Scaringe

By No Priors Podcast

Categories: AI, VC

Summary

Rivian's CEO believes that by 2030, every car will be expected to drive itself, as the industry shifts from rules-based to neural network-based autonomous architectures. The company is rebuilding its autonomy stack from scratch, seeing over 50x progress in the next 5 years compared to the last.

Key Takeaways

  1. Autonomy is a core part of Rivian's strategy, with a complete rebuild of their perception and planning stack to use neural networks instead of rules-based approaches.
  2. Rivian expects over 50x more progress in autonomous driving capabilities in the next 5 years compared to the previous 5 years, as the industry shifts to true AI architectures.
  3. Building a data flywheel is crucial for training autonomous models, requiring Rivian to grow its car fleet and data collection.
  4. The transition from rules-based to neural network-based autonomy architectures is difficult for incumbent automakers, as it requires discarding much of their prior work.
  5. Rivian is taking an in-house approach to developing its autonomy stack, rather than partnering, to maintain control over a key strategic capability.
  6. Rivian believes the world needs more EV choices beyond the Tesla Model Y, as consumers want variety and to self-identify with their vehicles.

Topics

Transcript Excerpt

By 2030, it'll be inconceivable to buy a car and not expect it to drive itself. Every single one of our cars, we want to have the ability for it to operate at very high levels of autonomy. Radars are extremely cheap. LARS are very cheap, but the really expensive part of the system is actually the onboard inference. In order to imagine more expensive than any of the perception stack, my view is EV adoption in the United States is a reflection of the lack of choice. As consumers, we need lots of c...