Nvidia CEO says company has achieved AGI, calls the milestone ‘senseless’

On an earnings call, Nvidia CEO Jensen Huang said the company has achieved artificial general intelligence (AGI), but immediately dismissed the milestone as ‘senseless.’ Huang did not provide a definition or benchmark for AGI, and reiterated prior comments that the term lacks consensus across the industry.

Huang’s remarks followed a question about OpenAI’s pursuit of AGI. He responded that, for many tasks, Nvidia could claim to have already achieved AGI, while adding that such milestones are arbitrary and lack meaningful significance. He also pointed to a shift toward AI systems that act as autonomous agents capable of recursive learning and self-improvement.

Industry definitions of AGI vary widely and often lack measurable thresholds. OpenAI’s charter defines AGI as ‘highly autonomous systems that outperform humans at most economically valuable work,’ a standard Huang and others have criticized as unworkable. OpenAI has also reportedly used a financially driven definition with Microsoft—systems capable of generating at least $100 billion in profits—though company leaders have acknowledged the term’s ambiguity.

Other tech leaders have similarly questioned the utility of AGI. Anthropic CEO Dario Amodei has called the term ‘imprecise’ and a ‘marketing term,’ while Meta, Microsoft, Amazon, and Google DeepMind have promoted alternative phrases like ‘personal superintelligence,’ ‘humanist superintelligence,’ ‘useful general intelligence,’ and the ‘foothills of the singularity.’ Despite the proliferation of new terminology, none has clarified what AGI actually means or how it would be measured.