Nvidia CEO Claims AGI Achievement While Dismissing Term as 'Senseless'

During a Wednesday earnings call, Nvidia CEO Jensen Huang announced that the company has "achieved AGI" for many tasks, as The Verge first reported. This follows a previous statement made by Huang in March during a Lex Fridman podcast where he stated, "I think we've achieved AGI."

Despite these assertions, Huang characterized such milestones as "senseless" during the earnings call. He did not provide a precise definition or a specific benchmark to quantify the achievement. Instead, Huang highlighted the shift toward autonomous agents capable of recursive self-improvement and learning new skills. He emphasized that the primary focus should be on AI performing "productive and useful work" and "generating profitable tokens," noting that increased compute leads to more tokens and higher profits.

The lack of a standardized definition for artificial general intelligence is a recurring theme across the industry. The Verge notes that while OpenAI defines AGI as systems that outperform humans at most economically valuable work, the company has reportedly worked with Microsoft on a different, financially-driven definition involving systems capable of generating at least $100 billion in profits. Other industry leaders have similarly questioned the terminology; Anthropic CEO Dario Amodei has described AGI as a "marketing term," while OpenAI CEO Sam Altman has admitted the phrase is "not a super useful term."