Nvidia CEO Reveals Next-Gen Tech at Computex 2025 | Fingerlakes1.com

At Computex 2025, Nvidia cemented its status at the epicenter of the artificial intelligence revolution, unveiling a wave of cutting-edge developments in humanoid robotics, AI infrastructure, and advanced chip technology.

While spotlighting global innovation, CEO Jensen Huang notably praised China’s AI researchers, calling them “world-class,” even amid ongoing U.S.-China tech tensions.

Key Highlights from Nvidia’s Computex 2025 Announcements

During his headline keynote in Taipei, Huang showcased a sweeping vision that touches nearly every frontier of artificial intelligence:

• Project GR00T (Generalist Robot 00 Technology):
  Nvidia introduced a new foundational model designed specifically for humanoid robots. Project GR00T aims to power general-purpose robots capable of understanding natural language and mimicking human movement. It’s part of Nvidia’s broader mission to build robotics platforms that can learn and adapt in real-world environments.
• Next-Generation Blackwell Platform Expansion:
  Huang detailed enhancements to the Blackwell GPU architecture, including new models such as the GB200 NVL72, a massive server rack built to support trillion-parameter AI models. This system uses 72 Blackwell GPUs and 36 Grace CPUs, connected with the ultra-fast NVLink switch fabric.
• Rubin AI Platform (Coming 2026):
  Nvidia teased its next-generation Rubin platform, set to succeed Blackwell, incorporating new AI chips, memory tech, and interconnects. The Rubin GPU will feature HBM4 memory, and its companion “Vera” CPU will enable even more powerful AI workloads.
• NIMs (Nvidia Inference Microservices):
  Huang also introduced AI “microservices” built for developers, allowing businesses to integrate pretrained AI models into cloud, PC, and data center applications with ease. These services span image generation, video, speech, and robotics tasks.

Nvidia’s AI Ambitions in Humanoid Robotics

The company is making a strong push into humanoid robotics, backing a range of startups—such as 1X Technologies, Agility Robotics, and Figure AI—who are building life-size robots. Each startup will integrate Nvidia’s Isaac robotics platform and the GR00T model, enabling the machines to perform complex physical and cognitive tasks.

“We believe every factory, warehouse, and home will one day have a robot,” Huang said, highlighting Nvidia’s belief in a “general-purpose” AI future beyond the cloud.

Despite trade barriers and U.S. restrictions on chip exports, Huang took time during his visit to praise China’s AI researchers as “world-class.” Speaking to Business Insider, Huang said:

“There’s no question that China is doing excellent work in AI. The researchers are top-notch.”

This acknowledgment signals Nvidia’s commitment to international collaboration—even as the U.S. government places increasing limitations on the company’s ability to sell its top-tier chips to China. Nvidia has reportedly designed new chips specifically for the Chinese market to comply with export rules.

What This Means for the Global AI Race

With Nvidia extending its leadership in GPUs, AI software, robotics, and now humanoid platforms, the company’s influence continues to stretch across industries—from self-driving cars and healthcare to defense and logistics.

• China’s AI sector is thriving independently, and Huang’s comments may help maintain open lines between researchers on both sides of the Pacific.
• The U.S. and global AI firms are under pressure to innovate or risk falling behind Nvidia’s rapid rollout of scalable platforms.

Why This Matters

Nvidia’s Computex 2025 announcements represent more than just product upgrades—they signal the next era of AI infrastructure. By launching models like GR00T and platforms like Rubin, Nvidia is shaping the future of robotics, cloud computing, and international AI collaboration.

As the line between human and machine capabilities continues to blur, Nvidia is ensuring it remains not just a participant, but the architect of that future.

MORE COVERAGE & UPDATES: Nvidia | Artificial Intelligence