Speaking on a Monday episode of the Lex Fridman podcast, Nvidia CEO Jensen Huang made a bold claim: “I think we’ve achieved AGI.”
Artificial general intelligence (AGI) is a loosely defined concept that has fueled intense debate among tech leaders, developers, and the public. It generally refers to AI systems that match or exceed human intelligence. In recent months, some industry figures have attempted to move away from the term, introducing alternative labels they argue are more precise and less overhyped, though these often describe similar capabilities. AGI has also become a critical benchmark in major industry agreements, including high-stakes contracts between companies such as OpenAI and Microsoft, where significant financial terms are tied to its definition.
Podcast host Lex Fridman describes AGI as an AI capable of effectively doing a person’s job - such as building and running a billion-dollar tech company. When asked whether AGI is five, 10, 15, or 20 years away, Nvidia CEO Jensen Huang replied, “I think it’s now. I think we’ve achieved AGI.”
Fridman noted that the claim would spark significant excitement. Huang pointed to the rapid rise of OpenClaw, an open-source AI agent platform, highlighting how users are deploying individual agents for a wide range of tasks. He added that it wouldn’t be surprising to see viral social applications emerge—such as digital influencers or interactive experiences that suddenly gain mass popularity.
However, Huang appeared to temper his earlier statement, acknowledging that many of these experiments fade after a short period. “A lot of people use it for a couple of months and it kind of dies away,” he said, adding that the likelihood of thousands of such agents collectively building a company like Nvidia remains effectively zero.
