Nvidia CEO Jensen Huang has hit back at skeptics who believed that the world didn't need Nvidia's expensive AI chips, following DeepSeek's recent success.
"Almost the entire world got it wrong," Huang said on stage at the conference, according to Reuters. He called the conference "the Super Bowl of AI."
"The amount of computation we need as a result of agentic AI, as a result of reasoning, is easily 100 times more than we thought we needed this time last year," he said, referring to autonomous AI agents that require little human intervention for routine tasks.
The need for Nvidia's expensive hardware was questioned a few months ago after DeepSeek's rise, as the Chinese AI company used older, less expensive AI chips to train its models. This also caused a shift as AI companies moved away from training their models on huge mountains of data to "inference" where the AI model uses its own intelligence to produce answers.
The bulk of Nvidia's $130.5 billion in sales last year was from the sale of its AI chips, some of which go for tens of thousands of dollars each, but despite Huang's presentation and vote of confidence, investors didn't seem impressed, as Nvidia shares tumbled 3.4% on Tuesday.
Huang also tackled more concerns about DeepSeek, and said that Nvidia's Dynamo open source software could make DeepSeek's R1 model dramatically faster.
"Dynamo can capture that benefit and deliver 30 times more performance in the same number of GPUs in the same architecture for reasoning models like DeepSeek," said Ian Buck, vice president and general manager of Nvidia’s hyperscale and HPC computing business.