menu
techminis

A naukri.com initiative

google-web-stories
Home

>

AR News

>

The high a...
source image

VentureBeat

1w

read

396

img
dot

Image Credit: VentureBeat

The high and low level context behind Nvidia CEO Jensen Huang’s GTC 2025 keynote | Dion Harris interview

  • Nvidia CEO Jensen Huang discussed high-level concepts and low-level tech details in his GTC 2025 keynote, teasing the imminent arrival of humanoid robots and self-driving cars.
  • Huang highlighted the shift from data-intensive retrieval-based computing to generative computing enabled by AI, focusing on synthetic data for rapid market entry of advanced technologies.
  • Dion Harris, Nvidia’s senior director, further elaborated on Huang's keynote, emphasizing advancements in AI models and the convergence of simulation, AI, and visualization in projects like Earth-2.
  • The concept of generative computing was discussed, indicating a shift towards combining retrieval-based elements with AI-assisted generation for streamlined design processes.
  • AI-driven workflows enable exploration of various design spaces and efficient decision-making processes, offering a new realm of possibilities.
  • Partnerships with various organizations and utilization of diverse data sources are crucial for the development of Earth-2, a platform aiming to create an intricate digital twin of the Earth.
  • The completion of Earth-2 involves stitching together different data sources through the OmniVerse platform, enhancing temporal and spatial consistency for a comprehensive Earth model.
  • Challenges in creating digital twins extend beyond scale, with complexities arising from modeling chaotic systems like the Earth or optimizing configurable spaces like factories.
  • Huang emphasized the significance of synthetic data, reasoning models, and inference for advancing robotics and autonomous machines, highlighting the importance of AI capabilities for real-world interactions.
  • Nvidia's Dynamo software plays a crucial role in deploying AI at scale, focusing on inference to bridge the gap between proof of concept and production for leveraging the full potential of AI.

Read Full Article

like

23 Likes

For uninterrupted reading, download the app