During a recent Monday episode of the Lex Fridman podcast, Nvidia CEO Jensen Huang made a controversial statement: “I think we’ve achieved AGI.” AGI, or artificial general intelligence, is a loosely-defined term sparking significant debate among tech CEOs, tech workers, and the public; it generally denotes AI that matches or surpasses human intelligence. In recent times, tech leaders have sought to distance themselves from this term, crafting new phrases that they believe are less exaggerated and more clearly defined, though these terms essentially mean the same as AGI. It’s also featured in significant contract clauses between companies like OpenAI and Microsoft, potentially influencing substantial financial stakes.
Fridman, the podcast host, defines AGI as an AI system that can effectively “do your job,” including starting, growing, and running a tech company worth over $1 billion. When asked when he thinks AGI will become real—whether in five, ten, fifteen, or twenty years—Huang said, “I think it’s now. I think we’ve achieved AGI.”
Fridman remarked, “You’re gonna get a lot of people excited with that statement.” Huang then talked about OpenClaw, an open-source AI agent platform, citing its viral success. He noted that individuals are employing their AI agents for various tasks and speculated about potential social applications or digital influencers that might suddenly become successful.
However, Huang appeared to retract his earlier statements slightly, acknowledging, “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 is zero percent.”