Artificial intelligence

Jensen Huang: The Future of AI Beyond Generative Models Toward 'Physical AI'

Jensen Huang, CEO of NVIDIA, emphasized that the next AI revolution will be 'Physical AI', which integrates digital capabilities with physical reality.

2026-11-16
2 min read
21 views
certi.news
Jensen Huang: The Future of AI Beyond Generative Models Toward 'Physical AI'

Jensen Huang, CEO of NVIDIA, explained that the next wave of innovation in artificial intelligence will not be limited to the digital capabilities seen in generative AI, but will transition to an advanced stage known as 'Physical AI'. This came during his participation in the NVIDIA AI Summit Japan, where he highlighted the radical transformation in how intelligent systems interact with the physical world around them.

The Difference Between Generative and Physical AI

While current generative artificial intelligence focuses on creating digital content such as texts, images, and computer code, physical artificial intelligence aims to enable machines and robots to understand the physical environment and interact with it with high precision. This means the transition of intelligence from merely processing data inside computer screens to the ability to perceive and make real-time decisions within factories, self-driving cars, and complex robotic systems.

What Changes Practically?

This transformation imposes new technical requirements that go beyond mere cloud processing power, as it requires:

  • Edge Computing Integration: To ensure ultra-fast machine response times when dealing with real-world variables.
  • Linking Software and Hardware: The need for AI models that understand the laws of physics and motion, not just linguistic patterns.
  • Developing Sensors and Processors: Increasing reliance on specialized chips that support immediate sensory perception.

Why Does This Trend Matter to Developers and Industry?

This trend represents a turning point in how autonomous systems are built. For developers, the challenge is no longer limited to training large language models (LLMs), but in building models capable of operating in unstructured and unpredictable environments. For industrial sectors, this shift means moving from traditional rule-based automation to smart automation capable of adapting and learning from the physical environment, opening unprecedented horizons in advanced robotics and smart logistics.

News source
MONOist Japan
Open original source ↗
c
Author

certi.news

Explore this story

Related topics and entities

In the same category

You may also like

View all news