When Making Mistakes Becomes Proof You’re Human
How AI is changing the way we recognize authenticity
By Wayne Meyers | Science and Society
I spent years believing good writing meant eliminating mistakes.
You revised. You polished. You fixed awkward sentences because you respected your reader. Good writing was supposed to feel effortless. The cleaner the prose, the more professional it appeared.
Now something strange seems to be happening online.
As AI-generated writing becomes more common, some people are intentionally leaving small imperfections behind. A typo here. An awkward phrase there. Maybe a sentence that feels slightly less polished than it could be. Not because they’ve forgotten how to write, but because they want readers to believe a real person is behind the words.
That shift may seem minor. But it points to a much larger question about authenticity, trust, and how we communicate in a world where machines can write convincingly.
What Changed?
The idea comes from an unusual trend that has started appearing in discussions about artificial intelligence and online communication. As AI writing tools become more capable, people are looking for new ways to determine whether content was written by a human being.
For generations, mistakes were treated as signs of carelessness. If a piece of writing contained spelling errors or awkward grammar, readers often assumed the author lacked attention to detail. The goal was always to remove those flaws before anyone saw them.
Today, that assumption is becoming less reliable.
AI systems can now produce writing that is grammatically correct, logically structured, and remarkably polished. In many cases, the output looks cleaner than what the average person writes. Ironically, that level of perfection can sometimes make readers suspicious.
As a result, some people have reportedly begun leaving small imperfections untouched. Others use more conversational language, personal anecdotes, or unusual turns of phrase. They’re trying to communicate something beyond information. They’re trying to communicate presence.
And presence is becoming harder to verify.
Why This Matters More Than Typos
At first glance, this seems like a trivial cultural quirk. But it may be one of the first visible signs of a much larger shift.
Technology has always changed how humans communicate. The printing press changed who could share ideas. Email changed the speed of communication. Social media changed who could build an audience.
Artificial intelligence may be changing something deeper.
For most of human history, creating content required effort that was difficult to fake. Writing a thoughtful article, drawing an illustration, composing music, or producing a video all required significant human labor. That effort became part of how we judged authenticity.
AI weakens that connection.
When machines can produce convincing work in seconds, readers begin looking for different signals. The question is no longer simply whether content is accurate or useful. Increasingly, people want to know who—or what—created it.
That’s a very different question.
The Society Question: What Changes for Humans?
If this trend continues, we may need to rethink how trust works online.
Today, most of us make countless judgments about authenticity without realizing it. We decide whether a review seems genuine. We decide whether a social media post feels sincere. We decide whether a writer sounds credible.
Many of those judgments rely on subtle human signals.
What happens when those signals become unreliable?
One possibility is that entirely new forms of verification emerge. Another is that audiences become less concerned about who created content and focus only on whether it is useful. A third possibility is that human identity itself becomes more valuable as AI-generated content becomes more common.
We’ve seen similar patterns before. Handmade products often become more desirable after mass production becomes widespread. Live performances gain value in a world filled with recordings. Scarcity changes perception.
Human-created communication may follow a similar path.
What We Don’t Know Yet
There are still more questions than answers.
- Will people eventually stop caring whether content is AI-assisted?
- Will new standards emerge that require disclosure?
- Will AI learn to imitate human imperfections just as effectively as it imitates grammar and structure?
- If nobody can reliably tell who created something, does authorship still matter?
The answer may shape far more than writing. It could influence education, journalism, marketing, entertainment, and nearly every profession built on communication.
The Bigger Picture
For centuries, technology removed friction.
Spellcheck removed spelling mistakes. Grammar software removed awkward sentences. Search engines removed the need to memorize information.
AI may remove effort itself.
But humans rarely abandon identity quietly.
If imperfections become valuable again, it won’t be because people forgot how to write. It will be because writing was never only about information. It was also about presence, personality, and connection.
The more machines learn to imitate us, the more we may find ourselves searching for evidence of one another.
And that may become one of the defining social questions of the AI era.
Science changes possibilities. Society decides what happens next.
What Do You Think?
If AI becomes impossible to distinguish from human communication, what signals will we use to recognize one another online?
Leave a comment below.
Enjoyed This Article?
Subscribe below to Science and Society for future-focused science, technology, and ideas that matter.
Based on reporting from Futurism regarding changing perceptions of authenticity in the age of AI-generated content.
