NVIDIA CEO Jensen Huang has outlined a bold vision for the future of AI, calling it ‘Physical AI’. This next wave of AI, according to Huang, presents a unique ‘three-computer problem’, or a three-body problem, requiring the integration of AI, robotics, and virtual worlds. This concept aligns with the growing trend of AI applications in the real world, moving beyond purely digital realms.
Huang highlighted the significant advancements in generative AI, with companies like OpenAI, Anthropic, Google, and Microsoft leading the way. However, he emphasized that ‘Physical AI’ represents the next frontier. This new form of AI will involve three distinct computer systems working in concert. The first computer will be responsible for creating the AI itself, utilizing the latest tools and techniques in machine learning. The second computer will act as a simulator, allowing the AI to be tested and refined in virtual environments using synthetic data generation. This is where NVIDIA’s Omniverse comes into play, providing a powerful platform for simulating complex real-world scenarios.
Finally, the third computer system will be responsible for running the AI in real-world applications. This will involve integrating the AI with physical robots, whether humanoids or specialized manipulators, allowing them to learn and adapt in real-time. The ‘three-computer problem’ highlights the complexity of bringing AI into the physical world, requiring a delicate balance between virtual simulation and real-world implementation.
Huang’s vision suggests a future where AI is no longer confined to digital spaces but interacts seamlessly with the physical world. This raises exciting possibilities for industries ranging from manufacturing and healthcare to transportation and customer service. NVIDIA, with its dominant position in AI hardware, is poised to play a pivotal role in shaping this future. As the company continues to innovate with its next-gen Blackwell AI GPUs, the journey towards ‘Physical AI’ is likely to accelerate.
