Humanlike thought five years off
Artificial general intelligence (AGI) could, by some definitions, arrive in as little as five years, Nvidia CEO Jensen Huang says.
Huang, who heads the world’s leading maker of artificial intelligence (AI) chips used to create systems such as OpenAI’s ChatGPT, was responding to a question at an economic forum held at Stanford University about how long it would take to achieve one of Silicon Valley’s long-held goals of creating computers that can think like humans.
Huang said that the answer largely depended on how the goal was defined. If the definition was the ability to pass human tests, Huang said, AGI would arrive soon.
“If I gave an AI … every single test that you can possibly imagine, you make that list of tests and put it in front of the computer science industry, and I’m guessing in five years we’ll do well on every single one,” said Huang, whose firm hit $2-trillion in market value on March 1.
AI can pass tests such as legal bar exams, but still struggles on specialised medical tests such as gastroenterology. But Huang said that in five years it should also be able to pass any of them.
But, by other definitions, Huang said, AGI may be much further away, because scientists still disagree on how to describe how human minds work.
“Therefore, it’s hard to achieve as an engineer because engineers need defined goals,” Huang said.
Huang also addressed a question about how many more chip factories, called “fabs” in the industry, are needed to support the expansion of the AI industry. Media reports have said that OpenAI CEO Sam Altman thinks many more fabs are needed.
Huang said that more will be needed, but each chip will also get better over time, which acts to limit the number of chips needed.
“We’re going to need more fabs. However, remember that we’re also improving the algorithms and the processing of [AI] tremendously over time,” Huang said.
“It’s not as if the efficiency of computing is what it is now, and therefore the demand is this much. I’m improving computing by a million times over 10 years.”