Newsletters
Nvidia CEO Jensen Huang said on Monday’s episode of the Lex Fridman podcast that AGI (artificial general intelligence), an AI term that matches or surpasses human abilities, has already been achieved.
His statement quickly spread through the tech industry and briefly pushed Nvidia shares up 1.5%.
Podcast host Lex Fridman framed artificial general intelligence as an AI capable of launching, scaling, and operating a billion-dollar tech company.
He asked Huang whether that benchmark was five, ten, fifteen, or twenty years away. Huang declined to offer a timeline, replying instead: “I think it’s now. I think we’ve achieved AGI.”
Huang’s statement came with an important qualifier. Moments later, he said, "A lot of people use it for a couple of months, and it kind of dies away. Now, the odds of 100,000 of those agents building Nvidia are zero percent."
This admission narrowed his definition of AGI to something more modest, an AI that could create a viral app, make money for a short time, and then fade away, rather than the world-changing version often discussed in public debates.
Huang referred to OpenAI's OpenClaw, the increasingly popular open-source AI agent platform, as proof that individual AI agents can already act independently and make an impact.
"I wouldn't be surprised if somebody created a digital influencer or some social application that feeds your little Tamagotchi or something like that, and it became, out of the blue, an instant success,” he said.
1 min read
Huang’s claim underscores a fundamental issue in the AGI debate: the term has varying meanings, and its definition shifts depending on the speaker.
At the 2023 New York Times DealBook Summit, Huang defined AGI as software capable of passing tests requiring human-level intelligence and suggested it could be achieved within five years. On the Fridman podcast, he used a much broader definition.
Other technology leaders hold differing views. OpenAI CEO Sam Altman told Forbes last month that his company had "basically built AGI, or very close to it," but later clarified this was meant in a "spiritual" sense and that AGI still requires "a lot of medium-sized breakthroughs."
Microsoft CEO Satya Nadella was more cautious, telling Forbes the industry is "not anywhere close" and that declaring AGI is "not about Sam [Altman] or me declaring it.”
Former Tesla AI chief Andrej Karpathy was even more circumspect with his estimation that AGI is still ten years away.
This debate has practical implications. AGI is now a key term in major business contracts, particularly between OpenAI and Microsoft, with substantial financial consequences depending on when and by whom the milestone is announced.
A United Nations report last year warned that AGI could "autonomously execute harmful actions beyond human oversight," posing risks ranging from threats to critical infrastructure to large-scale job losses.
Google DeepMind CEO Demis Hassabis has called for a U.N.-style organization to oversee AGI development.
Nvidia, whose Graphics Processing Unit (GPU) hardware powers much of the global AI boom, is now valued at approximately $4 trillion, making it one of the world’s most valuable companies.
Huang’s comments on the Fridman podcast, whether taken literally or not, have an impact beyond a single interview. The swift correction that followed may reveal as much about the current state of AGI as the claim itself.
Set as preferred source
Set as preferred source
1 min read
