Chip company NVIDIA wants to expand its leading role in technology for artificial intelligence applications with a new generation of its computing platform.
Company boss Jensen Huang presented the system, named Blackwell, at the company's own developer conference GTC in San Jose, California on Monday. "Blackwell is the engine to power this new industrial revolution" through AI, Huang said in his keynote address.
The system is touted to be four times more powerful than the current Grace Hopper generation when teaching artificial intelligence. NVIDIA's computer systems dominate AI training in data centres.
The company also wants to expand its role in the generation of content with the help of artificial intelligence. The Blackwell system is 30 times better at this than Hopper, Huang said. NVIDIA also has new software for this, which can also be used via interfaces in the cloud.
Huang was convinced that in the future, most content will not be prefabricated from storage, but that AI software will generate it freshly based on the current situation. NVIDIA has developed the computer system for this future.
With Grace Hopper, for example, the chatbot ChatGPT could have been trained within three months with 8,000 NVIDIA chips and a power consumption of 15 megawatts, said Huang. With Blackwell, this could be achieved in the same time with 2,000 chips and 4 megawatts of electricity.
The "AI superchip" is named after David Harold Blackwell, a University of California, Berkeley mathematician specializing in game theory and statistics.
NVIDIA also wants to promote the use of so-called "digital twins," in which companies can simulate their entire business digitally.
Before something is built in the real world, it will first be simulated digitally in future, Huang said. NVIDIA is also focussing on robots.
"Everything that moves will be robotic," Huang said. The aim is for robots to be able to learn simply by observing people. Huang said "the ChatGPT moment" for robotics could be just around the corner.