A fascinating debate began today when Nvidia CEO Jensen Huang said that we have already achieved AGI.

It sounds like a bold claim, but understanding what he actually means is essential.

According to Huang's definition, AGI is an AI capability that can perform functional-level economic tasks — like creating a service that generates a billion dollars in revenue.
He suggests that if an AI system can scale a large app with minimal human intervention, it is an example of functional AGI.
It’s no longer just a tool — it can make decisions on its own.

But this is where the debate heats up.
There is no universal definition of AGI.
Even scientists disagree on whether AGI has been achieved or not.
Many researchers say that today’s AI systems are still unreliable, cannot plan long-term, and struggle to understand the real world.
So, is Huang’s definition just a play on words?

I think it’s important to look at how quickly capabilities are advancing.
Whether you call it AGI or not, it’s clear that AI systems are now performing functionally complex tasks.
If it truly becomes AGI, software development, business operations, and the entire global economy will change.

For now, Huang’s comment raises a big question — has AI crossed a historic threshold, or is it just approaching one?
This debate isn’t over yet, but what’s clear is that functional AI capabilities are rapidly evolving, and that’s something everyone should pay attention to.