Why Rivian's RJ Scaringe is Betting on AI and Autonomy

Rivian founder and CEO RJ Scaringe is charting an ambitious course for the EV maker, built on deep control of hardware, software and AI.

From designing its own computer chips and sensors to building its own driving software and screens, RJ believes controlling all of the technology gives Rivian an edge and could one day lead to large-scale robotaxi deployment.

That vision was on full display at Rivian’s first “Autonomy and AI Day” in Palo Alto, where executives outlined a sweeping roadmap for automated driving and in-vehicle intelligence.

Custom silicon replaces NVIDIA

One of the most significant announcements was Rivian’s move to develop its own AI chip, replacing NVIDIA technology in future vehicles.

The chip will underpin Rivian’s next-generation autonomy platform and is designed to improve cost, performance and development speed.

RJ said in a statement: “Our updated hardware platform, which includes our in-house 1600 sparse TOPS inference chip, will enable us to achieve dramatic progress in self-driving to ultimately deliver on our goal of delivering L4.”

The chip, called the Rivian Autonomy Processor (RAP1), is a custom 5nm processor that integrates processing and memory onto a single multi-chip model.

RAP1 will power Rivian’s third-generation autonomy computer, ACM3, which the company says can process 5bn pixels per second. The hardware is expected to debut in R2 models beginning in late 2026.

Vidya Rajagopalan, Rivian’s Senior Vice President of Electrical Hardware, said: “With our in-house silicon development, we’re able to start our software development almost a year ahead of what we can do with supplier silicon.”

Cameras and radar development

Rivian also confirmed that future R2 vehicles will integrate lidar sensors alongside cameras and radar. Lidar uses laser light to measure distances, creating precise 3D maps of surroundings by timing how long it takes for light to reflect off objects and return.

While cameras remain central to Rivian’s approach, executives said recent improvements in lidar made the technology viable.

Vidya said: “Camera is the main workhorse of our sensor suite, generating the bulk of the data fed to the models, but the radar and lidar are critical to addressing the edge cases which would otherwise create the long tail of problem cases.”

A “large driving model” approach 

At the heart of Rivian’s autonomy strategy is its Large Driving Model (LDM), which the company described as being trained in a similar fashion to large language models.

RJ said: “With the deployment of our Gen 2 R1s, we began the process of building our data flywheel to grow and build our large driving model.

“Our approach to building self-driving is really designed around this data flywheel, where a deployed fleet has a carefully designed data policy that allows us to identify important and interesting events that we can use to train our large model offline before distilling the model back down into the vehicle.”

Rivian plans to expand hands-free assisted driving across more than 3.5m miles of roads in the US and Canada.

In 2026, it will introduce an autonomy subscription priced at either a US$2,500 one-time fee or US$49.99 per month.

RJ said Rivian’s third-generation autonomy platform will eventually support point-to-point driving and could progress toward eyes-off and fully autonomous operation.

“While our initial focus will be on personally owned vehicles, which today represent a vast majority of the miles driven in the US, this also enables us to pursue opportunities in the rideshare space,” the CEO added.