Rivian is considering a future in which drivers can use AI agents from other companies through the car’s software. CEO RJ Scaringe described that direction at TechCrunch Disrupt, but Rivian has not said when it might arrive.
From car commands to outside AI agents
Rivian recently introduced an AI assistant that lets people speak to the vehicle. Its current uses include adjusting the temperature, turning on seat warmers and summarizing the news.
The company says it built an in-house framework to interpret what a person wants and direct the request to the appropriate part of its software. Scaringe described the system as a “traffic light”: it can choose a model running locally in the vehicle or send a more complex question to the cloud, where it can use different models.
Scaringe said Rivian’s next step would be adding products from frontier AI companies such as Anthropic. He did not specify a launch date. When asked for more details, the company pointed to its “AI & Autonomy Day” event in December.
Connecting agents will take coordination
Rivian has developed its in-car software and many of its applications itself, including navigation. It has also integrated popular apps such as Apple Music. That gives the company a software environment of its own, but connecting outside AI products to it still requires technical work.
As of May of this year, Rivian’s agent could access external apps such as Google Calendar. Scaringe said broader integrations would require work behind the scenes and coordination with the AI labs to make the connections function.
That challenge includes deciding how an outside agent can use information collected in other apps. Scaringe imagines a driver asking the Rivian assistant to turn on ventilated seats, then continuing a conversation about planning a meal for guests. The car could draw on context from a conversation the driver had with an assistant on another device.
The example points to a more connected role for the car’s assistant: it could handle a vehicle request and help with a broader task in the same exchange. Making that possible depends on linking the car with an agent’s existing user context.
Vehicle controls stay out of an agent’s hands
Rivian says safety concerns shape how it plans to bring AI agents into the car. Scaringe said Rivian would not hand vehicle-level controls to an agent. An agent would not decide when to apply the brakes or steer.
That boundary separates conversational and app-based assistance from decisions that directly control the vehicle. Rivian’s approach, as Scaringe described it, is to use AI for requests and connected services while keeping safety-critical controls outside an agent’s authority.
A growing role for AI inside cars
Rivian would not be the first American car company to add AI agents to its software. Earlier this year, Tesla added a feature that lets users talk to SpaceX’s Grok chatbot and its Grok Bot assistant.
Rivian’s plans remain forward-looking: the company has discussed the direction and the work required, but has not announced when outside agents will be available. For now, its current assistant and app integrations offer a starting point for understanding how those services might fit into the vehicle.