The End of Human Exceptionalism? AI Wrote This

AI wrote this entire article.

Or did it?

Would you even know?

How would you determine if something was written by AI?

You might look for certain telltale signs—specific language patterns, formal structures, or other indicators. But that's only true if the person generating the AI text didn't put effort into refining it. If they don't care about the output, you might spot these signs. However, if someone is trying to manipulate you into believing the text is authentic, they can easily succeed. You won't be able to tell.

You see what I did there? Em dashes, — , is usually a sign of AI use. But I just told you that I’m the one writing this, not an LLM or other AI. Maybe you could run this through an AI detector to check for authenticity.

But here’s the thing: they're largely unreliable. They only work when the generator doesn't bother to make the text convincing. The truth is, there's no reliable way to distinguish AI-generated text from human-written text anymore. Even if you think you can be sure—maybe you know the author well or find the piece particularly evocative—you can't. With the right prompts and training, AI can fool anyone. Some people still deny this, but AI has advanced far beyond where it was just a couple of years ago. It's more capable than you probably realize.

So, did AI write this entire essay, or at least part(s) of it? The point is, you can't be sure.

I can assure you that I wrote this myself. I’ve been writing essays for awhile on my blog. My essays usually focus on difficult social issues and how they interact with popular culture. This sounds like an essay I would write, doesn’t it? Two years ago, it would have been obvious that I wrote this essay. But now? I can give you all the assurances I can, but the truth is, you can never be certain. This isn't a fad; it's reality.

We’re the only known species to have written anything down. We’ve recorded history with art and words for thousands and thousands of years. We’ve crafted online databases larger than could be imaginable just a few decades ago, recording everything known to us as a species. How do we know? Because it’s recorded. We can’t know anything else, for sure.

I recall very distinctly the first article I read in my adult life on AI and predictive text manufacturing. It was titled in the print edition of The New Yorker as, “Can a Machine Learn to Write for The New Yorker?” published in 2019.

Ah, 2019. Before OpenAI had hundreds of millions of users. Before AI was intruding into every application and piece of software that we use. Before the COVID-19 pandemic. Before masks and jabs. Before TikTok’s wild popularity that we see now. The good ol’ days.

Of course, 2019 was not that long ago, and yet, in terms of AI and language models, it might as well be forever ago. The aforementioned New Yorker article claimed that, at the time, the LLMs were “…like a person who speaks constantly but says almost nothing.” It was not predictive. ChatGPT, at the time in its GPT-2 variant, could not remember anything nor form its own thought. It could not remember anything it had said, it could only predict what was likely the next word based on the word at-hand.

By 2021, the same publication was stating, “GPT-3 hints at a world in which machines can generate language.”

Now, in 2025, its asking, “Why even try if you have AI?” and concluding, “If we’re not careful, then our minds will do less as computers do more, and we will be diminished as a result.”

I agree. We’re very much in danger here, and it’s no longer fair or wise to deny it.

There’s no undoing this technological progress, though. It’s here to stay and, most likely, to continue its expansion into every domain of our lives. In some versions of our world, perhaps that could be a good thing. No more email, no more shopping lists, no more wasted brain power on mundane everyday tasks, no more lost time to nonsense.

Unfortunately, that’s not how this is turning out. Instead, people aren’t being hired. Layoffs across tech are occurring, with expanding AI usage being cited as one of the reasons. People are losing their livelihoods. Children and teenagers aren’t learning. Critical thinking is being decimated. Artists and writers are being replaced.

So, did AI write this essay? Who fucking cares? Everything I write and think appears to be reproducible.

Header photo courtesy of Unsplash.