NOTE:
This article was written by ChatGPT 4.o with my expert supervision. It is an experiment, related to this article: “Can you tell if an article/video was AI-created?“
Could you really tell it was co-written by a robot?
The prompt I used was:
“Can you write a full article, make it awesome, with no need to copy my text fully, if it is not needed?”
You’re reading this article. Or maybe listening to it. Or watching someone read it to you.
Either way, the words are reaching your brain – and triggering thoughts.
Now pause for a moment.
Did a human write this? Or did an algorithm stitch it together in under a second?
In 2025, that’s no longer a hypothetical question. It’s a very real – and surprisingly difficult – one.
The rise of synthetic content
The age of AI-generated content isn’t coming. It’s here.
Text, videos, audio, images – entire websites, YouTube channels, and podcasts – are now produced with little to no human input. From blog posts and product reviews to motivational speeches and tutorial videos, machines are doing the talking.
And they’re good at it.
Disturbingly good.
YouTube: The illusion of presence
Let’s say you’re watching a YouTube video. The host is well-spoken. Articulate. Maybe a bit too polished. Are they reading an AI-written script? Using a teleprompter? Probably.
But here’s the kicker: they might not even exist.
In May 2025, Google demoed AI-generated actors having natural, unscripted conversations. The facial expressions? Spot on. The eye movement? Realistic. The dialogue? Better than many real creators manage. Watch the demo – and try not to feel weird about it.
Tools like Runway ML, CapCut, and Nvidia Broadcast now let anyone:
- Fake eye contact with the camera.
- Generate human-sounding speech.
- Animate realistic virtual avatars.
- Splice videos with zero technical skill.
What used to cost millions in post-production is now drag-and-drop.
Most of the Internet is already AI-generated
According to a 2025 study by Ahrefs, over 70% of new web pages are created by AI. And that’s just the detectable portion. The real figure? Likely higher.
On commercial websites, that number may exceed 95%. Corporate blogs, affiliate pages, e-commerce SEO filler – all mass-produced by machines trained on scraped human writing.
Let that sink in:
Most of what you now read online was not written by a person.
Can you tell the difference?
Sometimes, yes. Especially if you know the topic well. AI writing often feels… off. Too vague. Too confident. Weirdly repetitive. But not always.
Even experts get fooled.
And the average reader? They’re not just fooled – they don’t even know there’s something to question.
Studies from universities like UC San Diego show that large language models can now pass three-party Turing tests – convincingly impersonating humans in conversation. Reddit experiments have confirmed the same: people can’t tell they’re debating with a machine.
Meanwhile, some readers accuse actual human writers of being AI – just because the text is clean, structured, and typo-free. Irony, meet reality.
What this means for humans
If everything feels authentic, does authenticity still matter?
That’s the unsettling part. It’s not that AI is “tricking” us. It’s that we’ve stopped asking who – or what – is behind the content we consume.
We assume the video host is a person.
We assume the article has an author.
We assume the advice comes from experience.
Often, none of that is true anymore.
And when even AI systems are now trained on content written by other AIs, we enter a feedback loop. Machines learn from machines. Humans learn from machines. At some point, the original human input – the real stuff – gets diluted or lost.
So… does it matter?
AI isn’t inherently bad. Used properly, it’s just a tool – like a power drill.
It can save time, automate boring work, and enhance creativity.
But when it replaces actual thinking, judgment, or insight – when it floods the web with plausible-sounding nonsense – the consequences go beyond convenience.
- Experts get drowned out.
- Authentic creators get buried.
- Trust erodes.
- Learning becomes mimicry.
And worst of all: no one notices.
Final thoughts
The shift is already happening. We’re not looking at a future where AI-generated content exists.
We’re already living in a world where human-made content is the exception.
Will that change how we value knowledge, storytelling, or even identity online?
Maybe. Maybe not.
For now, just ask yourself:
How much of what I consume online was created by someone who actually exists – and knows what they’re talking about?
And if you’re not sure…
That’s the point.
Last updated:
Originally published:
