At Nvidia’s AI Developer Conference (GTC 2025), founder Jensen Huang announced significant advancements in AI and the future of the industry, reports Sarah Parvini of the Associated Press. He highlighted the surging demand for GPUs and predicted Nvidia’s data center revenue could reach $1 trillion by 2028. Huang unveiled new graphics architectures, Blackwell Ultra and Vera Rubin, and discussed AI’s evolution from perception to generative AI and, now, to agentic AI that can reason. He also introduced “physical AI” for robotics, emphasizing the role of synthetic data in AI training. Huang showcased Isaac GR00T N1, an open-source platform for developing humanoid robots, and announced collaborations with General Motors and Google DeepMind. Huang concluded by unveiling the “Halos” AI safety system for autonomous driving and introduced a new physics engine for robotics simulations, marking a major step toward generalist robotics. Parvini writes:
At GTC 2025 — dubbed the “Super Bowl of AI” — Huang focused his keynote on the company’s advancements in AI and his predictions for how the industry will move over the next few years. Demand for GPUs from the top four cloud service providers is surging, he said, adding that he expects Nvidia’s data center infrastructure revenue to hit $1 trillion by 2028.
Nvidia Isaac GR00T N1, is a customizable model for humanoid robots, designed to enhance reasoning and skills, reports Nvidia in a press release. Alongside this, the Isaac GR00T Blueprint enables synthetic data generation to train robots more efficiently. Nvidia also launched the Newton physics engine in collaboration with Google DeepMind and Disney Research to improve robot learning precision. These advancements aim to accelerate robotics development, address labor shortages, and create more adaptable humanoid robots.
