By Keerthi Ramesh
Nvidia is taking a major step toward reshaping how autonomous vehicles operate, unveiling a new artificial intelligence platform designed to help cars reason through unpredictable real-world situations rather than simply react to them.
At the CES technology conference on Monday, Nvidia CEO Jensen Huang introduced Alpamayo, a system the company says allows vehicles to interpret complex environments, handle rare driving events and clearly communicate why certain decisions are made on the road. The platform reflects Nvidia’s broader effort to move beyond chip manufacturing and become a foundational player in robotics and physical AI systems.
Huang said the technology was developed by training AI models directly on human driving behaviour, allowing machines to better anticipate risks and adapt to changing conditions. Unlike traditional driver assistance software, Alpamayo is designed to explain its choices in real time a feature Nvidia argues will be critical to building trust in fully autonomous vehicles.
READ: Nvidia takes $2 billion stake in chip designer Synopsys (
As part of the announcement, Nvidia revealed it has begun production on a driverless version of the Mercedes-Benz CLA, developed in partnership with the German automaker. The vehicle will debut in the United States in the coming months, with additional markets expected to follow later in Europe and Asia, according to the company.
During a live demonstration, a Mercedes equipped with Nvidia’s technology navigated city streets without human intervention while a passenger remained seated behind the wheel. Huang said projects like this have provided Nvidia with valuable insights into how artificial intelligence can be safely deployed in machines operating in the physical world.
Industry analysts said the move strengthens Nvidia’s position as competitors race to commercialize autonomous driving. Paolo Pescatore, an analyst at PP Foresight, said Nvidia’s ability to combine hardware, software and large-scale AI systems gives it a significant advantage in an increasingly crowded market.
READ: ByteDance to invest $14 billion in Nvidia (
Rather than locking the technology behind proprietary walls, Nvidia said Alpamayo will be released as an open-source model. The code will be available on the machine learning platform Hugging Face, allowing researchers and developers to adapt and retrain the system for their own autonomous vehicle projects.
The announcement also drew attention from rivals. Tesla CEO Elon Musk weighed in on social media platform X, saying: “I’m not losing any sleep about this. And I genuinely hope they succeed,” noting that while achieving near-perfect autonomy is possible, handling rare edge cases remains the industry’s biggest challenge.
Investor response was muted but positive, with Nvidia shares edging higher in after-hours trading. The company, currently the world’s most valuable publicly traded firm, also used the CES stage to confirm that its next-generation Rubin AI chips are already in production and scheduled for release later this year. Nvidia says the chips will deliver greater computing power while consuming less energy, potentially lowering costs for companies developing advanced AI systems.