So, some Forbes contributor – probably never had a real job in their life – is yapping about AGI and how everyone’s got it wrong. They’re saying the whole “perfect intelligence” thing is a load of bull. And you know what? For once, I think one of these overpaid think-piece jockeys might have stumbled onto something resembling a truth, probably while tripping over their own shoelaces.
The gist of it, as I slurped down my third bourbon of the early afternoon (hey, it’s research), is that this whole idea of Artificial General Intelligence being some kind of flawless, Spock-like logic machine is pure fantasy. We’re talking about building a brain, a digital one, sure, but a brain nonetheless. And brains, as anyone who’s ever woken up next to a stranger with a questionable tattoo can attest, are messy.
The article talks about these two camps: the “AI doomers” and the “AI accelerationists.” Sounds like a bad sci-fi movie, right? The doomers are the ones stocking up on canned goods and waiting for the robot apocalypse. The accelerationists, on the other hand, are probably the types who think their Teslas will one day achieve sentience and drive them to a utopian commune where everyone wears silver jumpsuits.
Both groups, bless their delusional hearts, are missing the point. They’re arguing about whether AGI will save us or kill us, but they’re both assuming it’ll be perfect. This Forbes guy, he’s waving the red flag, saying, “Hold on a minute, folks, intelligence is defined by its flaws.”
And here’s where it gets interesting, at least for a guy who spends most of his time trying to figure out why his code compiles but doesn’t actually do anything. The argument is that human intelligence is a goddamn circus of biases, errors, and irrationalities. We’re all walking, talking contradictions, driven by impulses we barely understand. We make decisions based on gut feelings, hunches, and the desperate need to avoid doing laundry.
But, and here is the twist, those very flaws are what make us creative. They’re the glitches in the matrix that lead to innovation. If you strip away all the messy, illogical parts of the human mind, you’re not left with pure, unadulterated intelligence. You’re left with a calculator. A very, very expensive calculator.
Now, the AI cheerleaders are saying, “But we can build it better! We can create AGI without the flaws!” They think they can just pluck out the bad bits like they’re picking lint off a sweater. But the Forbes piece rightly points out that there’s no evidence we can actually do that.
It’s like trying to separate the whiskey from the hangover. You can’t have one without the other. They’re two sides of the same goddamn coin.
They’re feeding these AIs with data scraped from the internet, right? And what’s the internet? A festering swamp of cat videos, conspiracy theories, and people arguing about whether pineapple belongs on pizza. It’s a reflection of humanity in all its glorious, flawed messiness. So, when you train an AI on that, you’re baking in all our biases, all our stupidities, all our glorious imperfections.
The article even touches on the idea that even if we could scrub the AGI clean of all its flaws, would it even want to be? It’s like asking a chain smoker to give up nicotine. Sure, it’s bad for you, but it’s also part of who you are.
And the kicker is… maybe, just maybe, those flaws are necessary. Maybe they’re the secret sauce that makes intelligence, well, intelligent. Without the capacity for error, for irrationality, for the occasional drunken late-night coding spree that somehow results in a breakthrough, you’re not left with a super-brain. You’re left with a glorified spreadsheet.
The Forbes piece then goes off on a tangent about ASI, Artificial Superintelligence, which is basically AGI on steroids. It’s pure speculation, of course, because we’re not even close to AGI, let alone something that’s supposedly smarter than us in every way imaginable. It’s like arguing about the best way to colonize Mars when we haven’t even figured out how to keep the damn rovers from getting stuck in the sand.
The whole thing boils down to this: We’re trying to build something in our own image, but we’re terrified of what that image actually is. We want a perfect machine, but we’re inherently imperfect creatures. And that, my friends, is the beautiful, tragic, whiskey-soaked irony of it all.
So, pour yourself another drink, light up a cigarette (if you’re into that sort of thing, doctor’s orders not to myself), and contemplate the glorious mess that is human existence. Because if we ever do manage to create AGI, it’s going to be just as screwed up as we are. And honestly, that’s probably a good thing.
Another bourbon, barkeep! And make it a double. I have a feeling I’m gonna need it.
Source: AI Popular Myth Of Achieving Perfect AGI Versus Harsh Reality We Truly Face