Rivian plans to run third-party AI agents in its cars, with details promised in December

Rivian recently rolled out an AI assistant that lets drivers and passengers interact with the company's electric vehicles just by talking to them. But founder and CEO RJ Scaringe said at TechCrunch Disrupt this week that, in the future, Rivian's cars will be able to run AI agents made by other companies, like Anthropic.

Scaringe did not say when Rivian would implement this, and the company declined to offer more details, pointing instead to the "AI & Autonomy Day" event it is hosting in December. For now, the public picture is a direction rather than a timetable.

It is not hard to imagine how Rivian might integrate something like Claude, OpenAI's Dots or Meta's Muse assistant into its software. The company has developed its in-car software from the ground up instead of leaning on Apple CarPlay or Android Auto, and it has built many of its own applications, such as navigation, though it has also worked to integrate popular in-car apps like Apple Music.

Rivian says it has created its own in-house agentic framework that understands what action a person wants to take when they talk to the vehicle and routes it to the right part of the car's software. Most of the time that involves what you would expect: asking the car to turn down the temperature, activate the seat warmers or summarize the news.

In Scaringe's telling, think of it as a traffic light. When you ask the vehicle something, it decides which model to use: a local model running on the edge can handle it, or a more complex question can go up to the cloud — and even there, Rivian uses different models for different tasks.

As of May, the agent could access external apps like Google Calendar. Scaringe said at Disrupt that the next step would be to add products from the frontier AI companies, though making them work in the car will not be easy. There is "a lot of plumbing behind the scenes" that has to be done, he said, and it is not just inside Rivian: it also requires a lot of coordination with the labs to make sure they enable that plumbing. Putting this kind of software near safety-critical systems means Rivian has to be extremely careful as well.

On safety, the line is clear. Scaringe said anything at the vehicle level is not being handed to an agent — you would not want an agent to determine "it's time to apply the brakes, or time to steer the vehicle." Instead, he thinks the implementation will look more like an extension of how people already use agentic apps, which would mean Rivian has to connect the vehicle to all the context apps like Muse collect on their users.

He offered an example: a driver could ask the Rivian assistant to turn on the ventilated seats and then, in the same breath, say they talked yesterday about a meal and are hosting guests on Friday, and ask for help planning what to cook. The car could then pull the driver's context from a conversation they had with the assistant on another device. That continuity of memory across devices and settings is what Scaringe called "magical."

Rivian would not be the first American car company to add AI agents to a car's software. Earlier this year, Tesla added a feature that let users talk to a Grok chatbot as well as its Grok Bot assistant. For Rivian, the real question is whether December's AI & Autonomy Day delivers a concrete product shape, and whether it can wire up frontier models without ever touching vehicle controls.