Nvidia Director General Jensen Huang, Tuesday at the GTC 2025 Company Conference stuck in tradition and was filled with notifications. But the company also retired into a small history lesson.
During the car part of his speech, Huang referred to Alexnet, a nervous network architecture that gained widespread attention in 2012 when he won a computer image recognition competition. Designed by computer scientist Alex Krizhevsky in collaboration with Ilya Sutskyr (who would continue to find Openai) and a scholar of Ai Geoffrey Hinton, Alexnet reached 84.7% accuracy in an academic competition called Imagnet.
The result of progress led to a revival of interest in deep learning, a subset of learning the machinery that uses nerve networks.
It turns out, Alexnet urged Nvidia to go “everyone in” in autonomous vehicles, in the way he says Huang.
“The moment I saw Alexnet – and we have worked on the computer’s vision for a long time – the moment I saw Alexnet was such an inspiring moment, such an exciting moment,” he said on stage. “This led us to decide to continue to build cars that drive cars. So we have worked on cars that drive cars for more than a decade. We build technology that almost every single car company being driven.”
Nvidia has marked partnerships with multiple vehicles, automobile suppliers and technology companies that develop autonomous vehicles. Its last, an extended GM collaboration, was announced this afternoon.
Vehicle manufacturers like Tesla and autonomous vehicle developers Wayve and Waymo use GPUs for databases. Other companies knock the NVIDIA OMNIVERSE product to build “digital twins” of factories to prove practically production processes and design vehicles. Meanwhile, Mercedes, Volvo, Toyota and Zoox have used the Nvidia Orin Computer-On-Chip computer system, which is based on the Nvidia Ampere Supercompeting architecture. Toyota and others are also using the operating system focused on the safety of Nvidia, Driveos.
Upper DNA: Nvidia is embedded in the automotive industry -and more specifically, automated direction -.
