'Excited to see Indian startups build the local AI economy': Nvidia CEO Jensen Huang

Nvidia CEO Jensen Huang on Wednesday said India has a major opportunity to build a local AI economy. 

"AI is moving fast, so every country has to move faster," Huang told reporters at the G20 Innovation Ministerial in Chapel Hill, North Carolina, when asked what his message would be for Prime Minister Narendra Modi on AI.

Huang said India's established information technology industry gives it a strong base to take advantage of the emerging AI platform.

"India is one of the world's leading IT industries, and this is the beginning of a new IT platform," he said. "I have the benefit of seeing Modi ji a decade ago talking about AI. I'm very excited to see local companies and startups starting in India starting to take advantage of the technology, contributing to the technology, and then building the local AI economy over."

Don't Miss: NVIDIA CEO Jensen Huang says AI data centers will unlock high-paying jobs beyond tech

Regulate Actual Harm

Asked whether the rules governing AI and innovation were in place, Huang said the technology was still in its formative stages. "The technology is in its formative stages, and the advice that I would have is to regulate practical and actual harm and not regulate theoretical and hypothetical harm," he said.

The Nvidia CEO said regulation should account for the fact that the technology was continuing to advance. "We want to make sure that the technology advances in a safe way and the more advanced it is, the safer and more useful it could be," he said.

Technology Must Become Safer

On whether the world needs international standards for AI safety, Huang said the priority should first be to make the technology itself safer. "The first thing we have to do is advance the technology so that it's safe," he said, comparing AI with cars, which have become safer as the underlying technology improved.

Huang said advances over the past few years had already reduced AI hallucinations and made systems more grounded in factual information and better at solving practical problems. "We need to keep moving in that direction," he said.

At the same time, Huang said companies could not wait for global standards to emerge and had to take responsibility for developing AI safely. "Meanwhile, every single company has to take it upon themselves to develop the technology safely," he said.