Nvidia's CEO Says AGI Is Here — And Also Meaningless

Nvidia's CEO Jensen Huang made a bold announcement on the company's earnings call: Nvidia has "achieved AGI." But then he immediately dismissed that achievement as "senseless." His point? Nobody actually agrees on what AGI means, so claiming to hit the target is kind of arbitrary. Huang says what really matters is whether AI is doing useful work and making money — not whether it meets some vague definition of "human-level intelligence."

The term AGI has always been fuzzy. OpenAI, the company founded to build it, defines AGI as "highly autonomous systems that outperform humans at most economically valuable work." But OpenAI privately uses another definition with Microsoft: AI that generates $100 billion in profits. Meanwhile, Meta talks about "personal superintelligence," Amazon says "useful general intelligence," and Google DeepMind prefers plain "general intelligence." Even Anthropic's CEO calls AGI a "marketing term." So when CEOs say they've reached AGI, they're really just making a statement about their own internal checklist.

Huang's bigger point is that the industry is moving beyond simple prompts toward AI agents that can learn new skills and improve themselves. Instead of chasing a milestone with no agreed definition, he says to focus on what AI actually does: generating "profitable tokens" as compute gets cheaper and more powerful. For everyday people, this means you should judge AI by its practical results — like saving you time at work, writing better emails, or helping you plan a trip — not by whether a company claims it's "AGI."

The catch is that without a shared yardstick, companies can declare victory whenever it suits them. Treat "AGI achieved" headlines with skepticism. What you can trust are concrete products you can test today. When the CEO of the world's most valuable chip company says the biggest buzzword in tech is "senseless," it's a reminder: don't get caught up in the hype. Pay attention to what the technology actually does for you.