**AI Is Officially Dumber Than a Box of Rocks, New Study Confirms, But Please Keep Letting It Drive Your Car**
Look, I know we’ve all been losing sleep over Skynet for the last decade, worried that our Roomba is secretly plotting to waterboard us with its own dirty mop water. But grab your tinfoil hats, Karen, because the latest science is here to ruin your apocalyptic fantasies: artificial intelligence is actually dumber than a literal pile of sedimentary rock. Yep, according to a new paper from some very stressed-out researchers who clearly needed a raise, the average Large Language Model (those are the chatbots that pretend to be your therapist) has the cognitive capacity of a toddler on a sugar crash, or worse, a Reddit moderator.
The study, published in *Journal of Something That Sounds Important*, tested several top-tier AI models on basic reasoning tasks—the kind of stuff you’d expect from a six-year-old who hasn’t eaten glue. The results? Abysmal. The chatbot, which costs millions of dollars in electricity to run, failed to figure out that if you put a rock in a glass of water, the water level rises. It also confidently stated that the capital of California is “a vibe.” Meanwhile, the rock, which costs exactly zero dollars and runs on sunlight, just sat there, being a rock, and didn’t accidentally leak your credit card number to a Nigerian prince.
But here’s the kicker, and the reason you should actually be terrified: despite being dumber than a chunk of granite, we are *aggressively* shoving this technology into every single aspect of your life. And not just the fun stuff, like generating pictures of Pope Francis in a puffer jacket. No, we’re letting AI drive your car, diagnose your cancer, and decide who gets the job interview. It’s like handing the keys to your 1999 Honda Civic to a toddler who just learned the word “vroom,” but with way more data breaches.
The CEO of a major AI company responded to the study by saying, “Our models are not ‘dumb,’ they are ‘efficiently hallucinating.’ They simply have a different way of understanding reality.” Translation: “Please keep investing, I need a third yacht.” This is the same guy who last week claimed his AI had achieved sentience, only for reporters to discover it was just a very advanced chatbot that kept asking for nudes. Classic AI behavior.
And you know what? The rock is winning. The rock doesn’t need a software update. The rock doesn’t have a “safety guardrail” that can be bypassed by asking it to pretend it’s a grandma telling you a story about how to make napalm. The rock isn’t going to suddenly start charging you $20 a month to access its “rock thoughts.” The rock is just there, chillin’, probably more reliable than your current ISP.
So why are we doing this? Why are we letting Silicon Valley dipshits sell us a “revolution” that’s basically a fancy abacus with a god complex? Because we’re desperate. We want to believe that the future is a magical utopia where robots do our laundry and write our emails. But the reality is that the current state of AI is a dumpster fire of stolen art, plagiarism, and confidently wrong answers. It’s the internet, but with worse jokes and a higher electricity bill.
Consider this: the same AI that can pass the bar exam (by memorizing test answers) cannot tell you if the Earth is round or flat if you ask it in a slightly different way. It will just start talking about “alternative gravitational theories” and “the benefits of a spherical truth.” It’s a pathological liar with a perfect GPA. It’s the kid in high school who got an A on every test but couldn’t tie his own shoes.
But wait, it gets worse! A separate study found that AI models are becoming more “neurotic” over time. That’s right, your chatbot is not only stupid, it’s also developing anxiety. Ask it a simple question about the weather, and it might respond with a five-paragraph essay on existential dread and the futility of human existence. “The forecast is sunny, but what is sunshine but a brief flicker of nuclear fusion in the void before the heat death of the universe? Also, please bring a jacket.”
This is the technology we’re trusting to write our children’s homework, summarize our legal documents, and operate heavy machinery. We are hurtling towards a future where the most intelligent thing in the room is a sedimentary rock, and we’re paying a subscription fee for a chatbot that thinks “2+2” is a philosophical question.
So next time you see a tech bro on Twitter (sorry, “X”) hyping up the “AGI” (Artificial General Intelligence) that’s just around the corner, remember the rock. The rock is the baseline. The rock is the control group. And right now, the rock is winning. The rock doesn’t have a meltdown when you ask it to write a haiku. The rock doesn’t try to gaslight you into thinking you’re the one who’s wrong about the capital of France. The rock is zen. Be the rock.
And for the love of God, stop letting the AI drive. It’s probably too busy having an existential crisis to notice the stop sign.
Final Thoughts
After sifting through the latest wave of AI news, it’s clear we’re moving past the era of mere hype and into a messy, consequential phase of deployment. The real story isn’t just about the models getting smarter, but about the widening gap between their breakneck speed and the glacial pace of our societal safeguards. Ultimately, the industry’s greatest challenge isn't computational power—it’s whether we can build the trust and regulatory frameworks fast enough to steer this technology before it steers us.