NVIDIA's CEO outlines two simultaneous platform shifts: the move to building applications on AI and a fundamental change in the computing stack from programming to training, driving the modernization of $10 trillion in infrastructure.
The company is heavily investing in and open-sourcing frontier AI models across various domains, including biology (La Proteina), physical AI (Cosmos), and autonomous driving (Alpamayo), positioning open models as a key driver of AI proliferation.
NVIDIA is expanding its full-stack strategy by deeply integrating its hardware and software into enterprise platforms like Palantir and ServiceNow, and into core industrial design tools from Cadence, Synopsys, and Siemens.
The next-generation Vera Rubin platform is announced, promising a 10x improvement in AI factory throughput (performance per watt) over the Blackwell generation, drastically reducing the cost and energy required to train and operate frontier AI models.
12 quotes
Concerns Raised
The immense energy and capital cost required to train and operate frontier AI models.
Ensuring the safety and reliability of autonomous physical AI systems like self-driving cars.
Opportunities Identified
Modernizing an estimated $10 trillion of computing infrastructure for the AI era.
The emergence of physical AI as the next major market, particularly in autonomous vehicles and robotics.
Democratization of AI capabilities through high-performance open-source models.
Revolutionizing industrial design and manufacturing by integrating AI into EDA and PLM software.