There used to be a joke in the technology community that “Artificial general intelligence (AGI) is always a decade away.” AGI was said to have arrived when the intelligence of a machine possessed the ability to understand or learn any intellectual task that a human being can.
The reason it was a joke was because AI always seemed to be a few years away from the next promised breakthrough.
It is now clear the joke has to be retired. Whatever your bar for AGI, you wouldn’t want to bet against it being here in the next decade.
I was in conversation with a wise friend who pointed out that today’s AI is already smarter than most people most of the time.
That’s an understatement. Whether it is ChatGPT or Claude, I’m just consistently blown away by their ability and how quickly they get better.
So I agree with his line of thinking – for all practical purposes, AGI is here. And considering these models are the dumbest they’ve ever been, they’re going to continue to get smarter and smarter.
That brings with it a collection of first, second, and third order consequences. In the coming weeks and months, we’ll dig into them. Most of these will be in equal parts wondrous, scary, and exciting.
But it starts with that simple and profound assertion – AGI is here. Now let’s get to figuring out what that means.
