The Rise of the Generic Human

For the past year, I’ve spent a lot of time teaching people how to humanize AI-generated content.

I don’t regret one minute of it.

When ChatGPT first exploded onto the scene, the problem was obvious. AI wrote competently, but it didn’t write convincingly. The prose was polished without being persuasive, articulate without being interesting, and fluent without giving you the sense that anyone had actually lived behind the words.

It could explain almost anything, but it struggled to make you feel that another human being had cared enough about the subject to write it.

The advice that followed was both necessary and effective. We learned to tell better stories. We brought back our opinions, our experiences, our imperfections. We stopped asking AI to produce finished work and started treating it more like a collaborator.

We became better editors. Better question askers. Better judges.

If you compare the average AI-assisted article today with one from eighteen months ago, the improvement is undeniable.

So why has something felt increasingly off kilter?

It wasn’t one article that made me notice it. It was dozens of them.

I’d finish reading a thoughtful piece on LinkedIn or Substack and think, “That was good.” Then I’d read another one. And another. Different authors. Different industries. Different topics.

After awhile they began to blend together in my memory.

It’s not like they were robotic or poorly written or anything like that.

It just felt like they all seemed to belong to the same person.

For weeks I couldn’t quite explain it. I kept assuming I was imagining things. Maybe I was simply reading too much AI-assisted content. Maybe the platforms themselves were rewarding a certain style. Maybe good online writing was naturally converging on the same rhythm.

Then I started paying closer attention.

The stories were different, but they served the same purpose.

The opinions were different, but they arrived with the same emotional temperature.

The vulnerability was different, but it appeared in remarkably similar places.

Even the imperfections felt strangely intentional.

It reminded me of something that happens in neighborhoods over time. Someone paints their front door red. It looks distinctive, elegant, original.

A few months later another neighbor does the same thing. Then another.

Eventually the entire street is filled with red front doors.

No individual homeowner copied everyone else. Each person simply adopted something that genuinely looked better than what came before.

The result, however, is not greater individuality. It’s convergence.

I think that’s what we’re beginning to see with AI-assisted writing.

For two years we’ve been learning how to escape the obvious signs of machine-generated prose. We’ve learned to avoid the clichés, soften the transitions, vary the rhythm, add stories, embrace imperfection, sound conversational, sound authentic.

None of that advice is wrong. Heck, I’ve given much of it myself.

But advice has an interesting life cycle.

At first it creates distinction. Then it eventually creates convention.

The very techniques that once separated human writing from AI writing gradually become the expected way to write. They stop functioning as expressions of individuality and start functioning as social norms.

Without realizing it, we’ve begun teaching one another what an authentic person is supposed to sound like.

And that’s where I think we’ve accidentally wandered into new territory.

We spent two years trying to make AI sound more human.

I think we’ve started creating a standardized version of what “human” looks like.

Not robotic or artificial or fake.

Just broadly recognizable.

The Generic Human.

The Generic Human is thoughtful. Curious. Warm. Self-aware. Willing to admit uncertainty. Comfortable telling personal stories. Respectful of nuance. Emotionally intelligent. Never excessively polished. Never excessively rough.

There is nothing objectionable about this person.

In fact, most of us would probably enjoy having coffee with them.

The problem isn’t that they’re unlikeable.

The problem is that there seems to be more and more of them.

Ironically, the more seriously we take authenticity as a technique, the easier it becomes to imitate.

That’s because authenticity is not actually a technique.

It’s a byproduct.

It’s what happens when thousands of invisible decisions are made by someone with a particular history, a certain set of frustrations, a unique sense of humor, a personal collection of obsessions, and a special way of making sense of the world.

Those decisions are where a voice comes from.

Oh sure, sentence length and contractions and shorter paragraphs and strategic vulnerability can express a voice.

But they cannot manufacture one.

That’s why I don’t think the next challenge is simply humanizing AI.

We’ve made a lot of progress there.

The next challenge is resisting the temptation to humanize ourselves into the same person.

Because your greatest competitive advantage was never that your writing sounded human. Millions of people “sound” human.

Your advantage has always been that your thinking doesn’t sound like anyone else’s.

The way you connect ideas.

The analogies you can’t help making.

The assumptions you question.

The things you notice that others pass by without a second thought.

The subjects you refuse to leave alone.

None of those qualities can be reduced to a checklist. They emerge through judgment. Through curiosity. Through lived experience. Through the long, often frustrating process of thinking your way toward something that didn’t exist before.

That’s why I believe the conversation is changing.

For the last two years, the question has been, “How do we make AI content sound more human?”

I think the better question now is, “How do we make sure humanized AI doesn’t start making us sound the same?”

Those are very different problems.

And I suspect the second one will define the next stage of creative work.

That’s why I’ve become increasingly convinced that the answer isn’t another prompt.

It’s a different way of thinking.

Over the past several months, I’ve been developing something I call Pretzeling. It’s based on a simple observation: the most original ideas rarely emerge from the shortest path between question and answer.

They emerge from the detours.

From exploring a possibility that doesn’t quite fit. From following an analogy a little farther than you planned. From holding two contradictory ideas in your mind long enough for something unexpected to happen. From refusing the first perfectly reasonable answer because you have a suspicion there’s a better one hiding underneath.

That’s the part AI can’t do for you.

Because those moments aren’t the product of prediction. They’re the product of judgment, curiosity, experience, and the wonderfully inefficient way human beings discover what they actually think.

Ironically, the more powerful AI becomes, the more valuable that kind of thinking becomes.

If everyone has access to the same extraordinary machine, then the real advantage shifts back to the person asking the questions, making the decisions, recognizing the unexpected connection, and knowing which path is worth following.

That’s what Pretzeling is really about.

And it’s exactly what I’ll be teaching in my live masterclass happening tomorrow called Pretzeling: Make Your AI Content Unmistakably Yours.

If you’ve ever looked at an AI-generated draft and thought, “It’s good... but it doesn’t quite feel like me,” this masterclass will show you why—and, more importantly, how to change that.

Sign up here

Because the future doesn’t belong to the people who become better at sounding human.

It belongs to the people whose thinking is impossible to imitate.