Nvidia Corp. (NVDA) isn't content being the AI chip king powering data centers. Now it wants to put its silicon inside every robotaxi on the road, and it's making moves that put it squarely in competition with Tesla Inc. (TSLA) and Alphabet Inc.'s (GOOGL) Waymo.

Level 4 Autonomy Coming in 2027, Nvidia Says

At CES 2026 on Monday, Nvidia announced it's working with robotaxi operators to get fleets on the road as early as 2027. Xinzhou Wu, Nvidia's vice president of automotive, said the company expects to enable "Level 4" robotaxis by that time. That's the technical term for vehicles that can operate without a human driver in defined geographic areas.

Here's the context: Nvidia has been selling automotive solutions since 2015, but it's still a relatively small slice of the business. CEO Jensen Huang, though, has identified robotics and self-driving technology as Nvidia's next major growth engine after AI infrastructure. Translation: this matters a lot more than the revenue numbers currently suggest.

The "ChatGPT Moment" for Cars and Robots

To make autonomous development easier and faster, Nvidia is rolling out new hardware and software. The Drive AGX Thor automotive computer costs roughly $3,500 per chip, which Nvidia says will lower research costs and speed up development timelines for automakers.

But the real headline is Alpamayo, an open-source family of Vision Language Action models that combine perception, reasoning, and action into one system. Think of it like giving a car the ability to see, think, and act all at once, rather than handling those tasks separately.

Major players are already circling. Lucid Group, Inc. (LCID), JLR, Uber Technologies, Inc. (UBER), and Berkeley DeepDrive are all exploring how to use Alpamayo to build autonomous vehicle systems capable of Level 4 autonomy.

Huang framed this as the "ChatGPT moment for physical AI," marking a turning point where machines can understand, reason, and act in real-world environments. He pointed to robotaxis as early beneficiaries because Alpamayo enables autonomous vehicles to reason through rare edge cases, operate safely in complex settings, and explain their driving decisions. Those capabilities, Huang said, are essential to building safe, scalable autonomy.

Taking Aim at Tesla and Waymo

Nvidia's robotaxi ambitions put it in direct competition with fleet-based operators like Alphabet's (GOOGL) Waymo. The company recently demonstrated its system in a 2026 Mercedes-Benz CLA driving through San Francisco. Nvidia currently classifies that system as "Level 2 Plus Plus," which is comparable to Tesla's Full Self-Driving.

Nvidia aims to reach point-to-point self-driving for consumer vehicles by 2028. That timeline puts pressure on Tesla, which is already facing mounting scrutiny over its autonomy claims. Ross Gerber of Gerber Kawasaki has said Tesla will not achieve Level 4 or Level 5 autonomy without addressing hardware limitations, not just software improvements.

Adding to the pressure, Tesla has revised its Full Self-Driving disclosures to clarify that the system does not provide autonomous capability and requires constant human supervision. That's a notable admission in a market where competitors are racing toward true driverless operation.

Why This Matters

The autonomous driving market is heating up fast, and Nvidia is positioning itself as the infrastructure provider for the entire industry. By offering chips, software, and open-source models, it's trying to become the default platform for robotaxis and autonomous vehicles, much like it became the default for AI training.

The question is whether automakers and robotaxi operators will embrace Nvidia's ecosystem or build their own. Tesla has famously gone its own way, developing its own chips and software. But for companies that don't want to reinvent the wheel, Nvidia's offering looks increasingly compelling, especially with a 2027 timeline for Level 4 robotaxis.

If Huang is right about this being the "ChatGPT moment" for physical AI, then the race for autonomous dominance is about to get a lot more interesting. And Nvidia just put itself right in the middle of it.

Price Action: Nvidia shares were up 0.49% at $189.04 during premarket trading on Tuesday.