AI agents are coming to Rivian’s cars

Key points

• Rivian CEO RJ Scaringe said future Rivian cars will run AI agents from companies like Anthropic.
• Scaringe gave no timeline and Rivian pointed to its December AI & Autonomy Day event.
• Vehicle-level controls such as braking and steering will not be handed to AI agents.

Rivian plans to let its cars run AI agents built by outside companies such as Anthropic, founder and CEO RJ Scaringe said this week at TechCrunch Disrupt. The carmaker recently launched an in-house AI assistant that lets drivers and passengers talk to its vehicles.

Scaringe did not give a date for adding third-party agents, and Rivian declined to share more details, pointing instead to its AI & Autonomy Day event in December.

How Rivian's assistant works

Rivian developed its in-car software itself rather than relying on Apple CarPlay or Android Auto, and built many of its own apps, including navigation. The company says it created an in-house agentic framework that identifies the action a person wants and routes it to the right part of the car's software.

Typical requests include lowering the temperature, turning on seat warmers or summarizing the news. As of May this year, the agent could access external apps such as Google Calendar.

You wouldn’t want an agent to determine, ‘hey, it’s time to apply the brakes, or time to steer the vehicle.’

— RJ Scaringe

Safety limits and next steps

Scaringe said adding products from frontier AI companies would require extensive behind-the-scenes work, including coordination with the labs. He said placing such software near safety-critical systems means Rivian must be extremely careful.

He said vehicle-level controls will not be given to an agent, so an agent would not decide when to brake or steer. He expects the feature to work more like an extension of current agentic apps, with the car pulling in context those apps collect, such as a conversation held on another device.