AI // NVIDIA CEO Jensen Huang
- Jensen Huang, the founder and CEO of Nvidia, discusses Nvidia’s focus over the next 10 years, particularly around scaling up computing power and performance through co-design of hardware and software.
- Nvidia is aiming to double or triple computing performance every year at scale, outpacing Moore’s Law through techniques like disaggregation, data center-scale computing, and tight integration of the hardware and software stack.
- A major focus is enabling extremely low-latency inference for large language models and other AI applications that require rapid token generation, while also maintaining high throughput for cost-effective training and inference.
- Nvidia is building fully integrated data center solutions that can be easily deployed and optimized, treating the data center as the fundamental unit of computing rather than individual chips or servers.
- The use of AI and machine learning is transforming every field of science and engineering, with a coming “tidal wave” of AI-powered breakthroughs across disciplines. Huang sees AI as a universal enabler that will drive discovery and progress in unexpected ways.
- Personally, Huang makes heavy use of AI tools like ChatGPT as a tutor and constant reference, double-checking even his own expertise against AI-generated responses.
