NVIDIA Sets Sights on Robotics Revolution with Jetson Thor Chip Launch
NVIDIA, amid intensifying competition in the AI chip sector, is shifting its focus towards robotics as the next significant growth avenue. The tech giant plans to introduce its latest humanoid robot chip, Jetson Thor, in the first half of 2025. This strategic move marks NVIDIA's commitment to establishing itself as a leader in the budding robotics revolution by selling a comprehensive "full-stack" solution that integrates AI training software with onboard chips.
Deepu Talla, NVIDIA's Vice President of Robotics, expresses anticipation about the impending "ChatGPT moment" for physical AI and robotics, highlighting a market approaching a critical juncture. He attributes this shift to two pivotal technological advancements: the prolific growth of generative AI models and the capability to train these models in simulated environments. This latter development is essential for addressing the "sim-to-real gap," ensuring robots trained in virtual simulations can operate effectively in the real world.
Although NVIDIA has not disclosed specific sales figures for its robotics products, these offerings currently constitute a minor portion of its total revenue. For instance, data center revenue accounted for approximately 88% of NVIDIA's $35.1 billion sales in the third quarter. Nonetheless, the robotics market transformation is being hastened by technologies that allow simulation-based training, enhancing the robots' real-world performance.
In past months, NVIDIA has gradually been laying the groundwork for its robotics strategy. As far back as March, the company's CEO showcased various robots powered by NVIDIA chips. Rather than directly competing with manufacturers like Tesla, NVIDIA aims to serve as a foundational OEM to a multitude of global robotics makers, presenting its platform as an investment in the coming wave of physical AI.
This strategic endeavor aligns with NVIDIA’s comprehensive approach, encapsulating development stages from AI training systems like DGX, simulation environments such as Isaac Lab, and the processing power of the Jetson Thor. In collaboration with companies leveraging its technology, NVIDIA is striving to bridge the current challenges in robotics, including scaling issues and cost reduction, ultimately pushing the market toward its potential breakthrough.
