After 25, ChatGPT may quietly rob you of common sense—here’s why

The World Before and After ChatGPT

Just a few years ago, when Futura was celebrating its 25th anniversary (and journalists mulled over the most striking events and images of their decades in science), the world looked pretty different. On November 30, 2022, OpenAI launched ChatGPT and, like some blockbuster tech debut from Silicon Valley, the impact was immediate and global. Artificial intelligence, once a techie’s playground, became a term overheard everywhere from dinner tables to government offices. In less than four years, large language models—the very technology powering chatbots—became almost synonymous with the concept of AI itself.

But here’s the kicker: artificial intelligence was everywhere long before ChatGPT, just hiding in plain sight, and in forms very different from today’s generative AIs.

From Invisible Algorithms to Everyday Conversations

Before ChatGPT, AI mostly worked behind the scenes. It handled tasks, not chit-chat. Think recommendation systems on Amazon or Netflix, translation tools, spam filters, route planning, and even sorting your digital photo library. We developed a broad range of artificial intelligence systems—most of which were out of reach for anyone but researchers.

AI’s story goes much farther back, too. In 1943, Warren McCulloch and Walter Pitts built the first neural network. In 1950, Alan Turing proposed his famous test. The very phrase “artificial intelligence” is credited to John McCarthy during the Dartmouth Conference in 1956. Across popular culture, AI has popped up in classics like HAL 9000 (also called CARL 500 in some translations) from “2001: A Space Odyssey,” which hit theaters in 1968. But for everyday people, AI didn’t truly become real until ChatGPT arrived—you could finally talk to it, one-on-one.

“We are in the Singularity”—And the Rise of Cognitive Delegation

“We are in the singularity.”

That’s what Sam Altman, OpenAI’s chief, recently proclaimed—a phrase that’s sent shockwaves through the AI community. For Altman, the border between artificial and human intelligence is already beginning to blur. He claims that the technological singularity has begun and announces a historic turning point.

What does this look like in daily life? Here’s a very modern parable: someone I know, a well-educated adult, decided to roast a chicken. She asked a chatbot for the recipe—which, unfortunately, was full of mistakes. Even when another, more culinary-savvy person offered advice, she chose to trust the AI’s guidance. When she later asked how to check if it was cooked, the chatbot suggested using a probe to check the temperature. Taking the instructions literally, she placed it in the chicken’s cavity rather than in the meat—which (not surprisingly) resulted in a reading far too low. Luckily, the other person stepped in and saved the meal before we all risked undercooked chicken, but the anecdote illustrates a growing phenomenon: cognitive delegation. By offloading all decisions to AI, we sometimes stop using our own common sense—and start acting, well, a bit foolish. It’s no different than those infamous GPS stories, where drivers who follow directions blindly end up in the middle of a lake.

AI: A Blessing, a Crutch, and Maybe a Leveler?

There’s another risk to AI’s ubiquity: the risk of dumbing things down. Sure, AI can help boost one person’s creativity, but at the scale of humanity, it may actually water it down. Even seemingly harmless tools, like the recently launched ‘AI Overview’ in Google searches in France, can have an impact. For example, if you ask for a synonym, the AI answers right away—you don’t even need to look at the search results anymore.

Every time you let the AI read for you, here’s what you’re missing:

• Reading is a millennia-old act, and in recent decades, it’s been on the decline.
• The rise of AI that can summarize, analyze, or compare for us is accelerating that trend.
• But, is this amazing machine making us lose something essential?
• Reading is more than compiling a stream of data—it’s an intimate learning experience that awakens our awareness.
• What do we stand to gain (or lose) by letting AI take this over for us?

There’s one more twist. Language models come with their own biases and, over time, could even change the way we express ourselves—nudging us toward certain words, expressions, or sentence structures, even if we don’t interact with chatbots directly. This applies to every domain. For example, young artists who grow up surrounded by a sea of AI-generated content might absorb influences without ever realizing it. And then there are the hordes of bots polluting social media—sharing articles or leaving heavily biased comments to sway public opinion.

The Road Ahead: Bright Future, Blurry Hazards

The arrival of ChatGPT marked a real technological and societal turning point. It’ll likely take years—maybe even decades—before we can fully measure its consequences and decide whether the benefits truly outweigh the drawbacks. Until then, maybe it’s worth keeping our common sense on speed dial, just in case our GPS tries to direct us into a lake—again.