In keeping with tradition, Nvidia CEO Jensen Huang’s keynote address at the company’s GTC 2025 conference on Tuesday was jam-packed with announcements. However, the business also slipped in a little history lesson.
Huang mentioned AlexNet, a neural network architecture that became well-known in 2012 after winning a computer image-recognition competition, during the section of his lecture about automobiles. In an academic competition dubbed ImageNET, AlexNet, which was created by computer scientist Alex Krizhevsky in partnership with AI researcher Geoffrey Hinton and Ilya Sutskever (who would later start OpenAI), attained an accuracy of 84.7%.
The innovative outcome sparked renewed interest in deep learning, a kind of machine learning that makes use of neural networks.
As it happens, AlexNet encouraged Nvidia to “go all in” on driverless cars.the narrative style of Huang.
He remarked onstage, “It was such an inspiring moment, such an exciting moment, the moment I saw AlexNet — and we’ve been working on computer vision for a long time.” We made the decision to fully invest in creating self-driving cars as a result of that. For more than ten years, we have been developing self-driving automobiles. Our technology is used by practically all self-driving car manufacturers.
Numerous automakers, auto suppliers, and tech firms working on autonomous vehicles have partnered with Nvidia. Its most recent announcement this afternoon was an extended partnership with GM.Nvidia GPUs are used in data centers by automakers like as Tesla and autonomous car firms Wayve and Waymo. Other businesses use Nvidia’s Omniverse software to create “digital twins” of facilities in order to develop cars and test manufacturing procedures online. In the meanwhile, Nvidia’s Drive Orin computer system-on-chip, which is built on the chipmaker’s Nvidia Ampere supercomputing architecture, has been utilized by Mercedes, Volvo, Toyota, and Zoox. Nvidia’s safety-focused operating system, DriveOS, is also being adopted by Toyota and other companies.