NVIDIA is accelerating its strategic deployment in the autonomous driving sector by expanding collaborations with transportation giants such as Uber and Lyft, aiming to advance the commercial implementation of robotaxis.

During the recent GTC conference, NVIDIA announced plans to broaden its cooperative initiatives with companies including Uber and Lyft. This news provided a significant boost to their stock prices: Uber shares surged over 5% in recent trading, while Lyft's stock rose approximately 3%. Following a 1.6% increase on Monday, NVIDIA's stock experienced a slight decline of about 0.4% in recent sessions.

NVIDIA and Uber intend to launch autonomous vehicles equipped with NVIDIA's software in Los Angeles and San Francisco during the first half of 2027, with plans to expand the service to dozens of cities by the end of 2028.

Additionally, NVIDIA disclosed new or expanded partnerships with automotive manufacturers including Hyundai, BYD, Geely, Isuzu, and Nissan. The stocks of these companies also saw gains in their respective local exchanges.

NVIDIA's core technological advantage in autonomous driving lies in its newly open-sourced reasoning VLA (Vision-Language-Action) model, Alpamayo 1. This model is designed to enable vehicles to "think" and devise solutions in unexpected situations.

The company's CEO, Jensen Huang, stated: "The ChatGPT moment for physical AI has arrived—machines are beginning to understand, reason, and act in the real world. Autonomous taxis are among the first beneficiaries. Alpamayo equips self-driving cars with reasoning capabilities, allowing them to contemplate rare scenarios, navigate safely in complex environments, and explain their driving decisions—fundamental requirements for safe and scalable autonomous driving."

The first vehicles incorporating NVIDIA's technology are expected to hit roads in the United States during the first quarter of this year, followed by Europe in the second quarter and Asia in the latter half of the year.

Unlike traditional models that directly map visual inputs to actions, the reasoning VLA model can decompose complex driving tasks into manageable sub-problems and present its reasoning process in an interpretable manner. For instance, when approaching an intersection, the system engages in human-like reasoning: "I see a stop sign, a vehicle approaching from the left, and pedestrians crossing the street. I should slow down, come to a complete stop, wait for the pedestrians to cross safely, and then proceed when it is safe." This capability is crucial for addressing rare and unpredictable "long-tail" scenarios in autonomous driving.

Sarfraz Maredia, Uber's Global Head of Automated Mobility and Delivery, commented: "Handling long-tail and unpredictable driving scenarios is one of the defining challenges for autonomous driving. Alpamayo creates exciting new opportunities for the industry."

Beyond autonomous vehicles, NVIDIA is also expanding its footprint in the broader field of physical AI. The company has released several open-source models and tools, including the Nemotron family for agent AI, the Cosmos platform for physical AI, Isaac GR00T for robotics, and Clara for biomedical applications.