The truth is, large language models like ChatGPT, Gemini, Claude, and Grok have all been casually dunking on the Turing test for a while now. Most people can’t tell if they’re talking to a human or an AI anymore - and they’ve kind of stopped trying. You read a tweet, a comment, an entire article, and your brain quietly wonders, “Hmm… was this written by a person?” And then it just moves on. We all do. The game is over, but no one seemed to care that we won (Or the computers won. depends how you look at it).
At first, this bothered me. A lot. I had been waiting for this moment for decades. This was supposed to be the moment. The event that would change everything. If we ever figured out teleportation or achieved nuclear fusion at room temperature, would that also just... casually drop on a Wednesday and get buried under a new trailer for Deadpool 4?
But the more I thought about it, the more I realized: the problem isn’t with us. The problem is with the test itself.
The Turing test, like most things invented in the 1950s, sounds cooler than it actually is. Alan Turing, genius that he was, imagined a future where we’d be able to “talk” to computers via a typewriter and a screen. His version of AI was based entirely on one thing: language. And the test he came up with wasn’t even a test so much as a vibe check. “If you chat with something and can’t tell whether it’s human or not, then that something is intelligent.” That was it. No scoreboard. No metrics. No criteria. Just a general idea of fooling the average human.
And that’s the problem. What human? A trained interrogator? A grandma from Ohio? How long do they have to be fooled for? Do they get follow-up questions? What if the human thinks another human is a bot? (Which happens. A lot.) Did the bot still pass? What if it answers convincingly nine times out of ten, but then tells you that giraffes are birds? Is it still intelligent? Or just glitchy?
See, the Turing test sounds like a scientific benchmark, but it’s really more of a philosophical thought experiment. It doesn’t measure comprehension or reasoning or truth. It measures how good a machine is at mimicking us. That’s it.
And LLMs? They are very good at mimicking us.
So yes, if your benchmark for “true AI” was “Can it hold a human-like conversation without giving itself away?” then congratulations, we’re already living in the future. But that doesn’t mean these models actually understand what they’re saying. It doesn’t mean they can think, reason, or make decisions. It just means they’ve gotten really good at doing an impression of someone who can.
It’s kind of like watching a parrot recite Shakespeare. Impressive? Absolutely. But that doesn’t mean the parrot understands death or love.
Modern AI doesn’t bother with the Turing test anymore. We’ve moved on. These days, we measure things like problem-solving, coding accuracy, logic, multi-step reasoning, creativity, and memory. LLMs get graded like overworked college students in a science fair. We want numbers. We want comparisons. We want models that can actually do things, not just talk a good game.
In hindsight, the Turing test feels kind of quaint. A relic from a time before computers could dream up photorealistic images of the pope in a puffer jacket. Before AI could write novels, pass the bar exam, or gaslight you into thinking ‘Em Dashes’ were always a ting. Turing imagined a future where conversation was the ultimate marker of intelligence. But we’ve learned that language alone isn’t enough. As Qui-Gon Jinn once said to Jar Jar Binks, “The ability to speak does not make you intelligent.”
So yeah, we found the holy grail. And nobody cared. Maybe because the grail turned out to be a little plastic cup labeled “I can do a pretty good human impression.”
But here’s the silver lining: now that we’ve crossed this weird, underwhelming milestone, we can finally ask bigger questions. Not “Can machines fool us?” but “Can they help us?” Not “Do they sound smart?” but “Can they be wise?”
Because if we’re going to live in a world filled with talking computers, I’d at least like them to say something worth listening to.
