AI content is crap. Here’s why

We asked ChatGPT:

AI content is crap. Answer why in 50 words

GPT’s response:

AI content often feels soulless and generic because it prioritizes efficiency over originality. It mimics patterns instead of creating meaning, lacks genuine emotion or lived experience, and recycles clichés. Without human insight or creativity, AI-generated writing becomes formulaic filler — optimized for algorithms, not authentic human connection or thought.

What an incredible revelation that is. And from the horse’s mouth, no less.

But wait. Let’s ask it the opposite:

AI content is great. Answer why in 50 words

GPT’s response:

AI content is great because it’s fast, scalable, and adaptable. It can generate ideas, summarize complex topics, and support creativity across languages and formats. By automating routine writing, AI frees humans to focus on strategy, storytelling, and innovation — making communication more efficient, accessible, and personalized than ever before.

Now we’re getting somewhere.

Let’s ask it to clarify:

So in a nutshell, read between the lines, it’s crap and good?

Exactly — in a nutshell, AI content is both crap and great. It’s crap at genuine creativity, emotion, and originality — it can’t truly feel or innovate. But it’s great at speed, scale, and structure — producing clear, consistent, and data-driven writing fast. In short: efficient, not inspired.

“Efficient, not inspired”

“Inspired” isn’t about artistic merit. It’s about whether the content does its job. Everything else is just publishing words into the void.

Here’s what AI can’t do for you:

• Replace real-world experience. AI has no concept of what happens when content meets readers in practice.
• Predict misunderstanding. AI can’t tell how your readers will interpret phrasing based on their context and expectations.
• Judge what persuades. It has no way to know what resonates with your audience versus what sounds reasonable.
• Challenge your brief. It executes instructions without questioning whether they’ll work in practice.
• Weigh competing priorities. Without explicit ranking, AI can’t decide what matters most.
• Question the premise. AI answers what you asked, not what you should have asked.

You can produce ten blog posts in an hour with AI. If none of them convert, you’ve spent an hour creating nothing of value. One piece of content that persuades someone to buy, subscribe, or reach out justifies days of work.

The calculation isn’t complicated. Content exists to change behaviour. Generic explanations don’t change behaviour. Insight does. Connection does. Understanding the reader well enough to address their hesitation, not just their stated question.

You can’t teach AI to replace a human

You can give AI more information, better prompts, and refined instructions. What you can’t give it is the ability to know when the information, prompts, and instructions are wrong and in what context.

Getting AI to write better requires you already knowing what better looks like. If you have that expertise, you’re not saving time but are only translating your knowledge into prompts. If you don’t have that expertise, you’re hoping AI will fill a gap you can’t even identify.

A human writer learns by seeing what works and what fails. They build intuition from consequences. AI has no concept of consequence. It cannot learn that last month’s content didn’t resonate, or that this approach burned your audience before, or that your competitor tried this strategy and it backfired.

Teaching implies learning from outcomes. AI can’t do that. It doesn’t know if last month’s content failed or why your audience stopped engaging. It just waits for your next prompt.

This is why businesses hire writers

Content writing agencies exist because producing content that works requires more than following instructions. It requires knowing what instructions to ignore.

We’ve spent 15 years watching what converts and what wastes budget. We know which content approaches work in your industry because we’ve published hundreds of pieces and seen the results.

When a brief asks for something that won’t land with your audience, we say so. When a topic needs reframing to actually address what your customers care about, we do it.

AI can’t do any of that. It executes. We strategise, challenge, and refine until the content achieves what you’re trying to accomplish.

Where AI belongs, where it doesn’t

Use AI for internal documentation, brainstorming, summarising research. Fine. But anything your customers read, anything that needs to persuade, inform, or convert? That requires judgment AI doesn’t have.

In Q1-Q3 2025, 87% of the articles we produced hit page one of Google. Our clients don’t have massively authoritative sites. The content works because it’s written by people who understand what readers and search engines respond to.

That’s our case. Now here’s what AI produces when asked to write the same article:

AI Content Is Crap: Here’s Why

by ChatGPT

In the past few years, artificial intelligence has transformed from a futuristic curiosity into a mass-production engine for text, images, music, and even code. Scroll through any social feed, marketing blog, or Amazon book listing, and you’ll find a flood of AI-generated “content” — articles, product descriptions, social captions, and even poetry. But much of it feels… off. Bland. Hollow. Forgettable. The truth is simple: most AI content is crap. Here’s why.

1. It’s Designed to Imitate, Not to Think

At its core, an AI writing model doesn’t think — it predicts. It doesn’t understand an idea, an audience, or a purpose. Instead, it statistically estimates which words are likely to follow others, based on patterns in the massive pile of text it was trained on.

That means AI writing is inherently reactive, never creative. It mirrors what already exists rather than inventing something new. When you ask an AI to “write a thought-provoking essay,” it pulls from the average of everything humans have written about essays and thought-provoking topics — the linguistic equivalent of microwaving leftovers. It can sound smooth, even insightful on the surface, but underneath there’s no original thought or lived experience holding it together.

2. It Lacks Voice, and Voice Is Everything

Great writing isn’t just about correct grammar or clear sentences. It’s about voice — that mix of rhythm, tone, and perspective that makes a writer sound unmistakably human. Think of Joan Didion’s sharp introspection, or Ta-Nehisi Coates’ moral gravity, or the crackling humor of Nora Ephron. You can recognize them instantly because their writing carries a pulse.

AI has no pulse. Even when it imitates style, it doesn’t feel it. The cadence may sound close, but it’s hollow — a mask with no face behind it. Because AI can’t experience fear, awe, heartbreak, or irony, it can’t anchor a sentence in emotion. The result is writing that’s technically correct but spiritually vacant. It’s the literary equivalent of elevator music: smooth enough to fill silence, but not enough to make you stop and listen.

3. It Flattens Human Creativity

AI content creation has a democratizing allure: anyone can make something quickly, with no cost, no gatekeeping, and no expertise. But that convenience comes at a price. When everyone uses the same tools trained on the same data, output starts to look, sound, and feel the same. The quirks, flaws, and wild risks that make human creativity vibrant are ironed out in favor of “good enough.”

We’ve already seen this in SEO-driven writing, where articles are churned out to satisfy algorithms rather than readers. AI simply accelerates the trend. Instead of nurturing originality, it rewards predictability. As more creators outsource their words to machines, the internet fills with echoing versions of the same 10 ideas, dressed in slightly different phrasing. The more AI writes, the more it feeds on itself, learning from its own regurgitated output — a loop of decaying originality.

4. It Can’t Replace Lived Experience

Writing isn’t just about stringing words together; it’s about filtering the world through perception. A war correspondent’s paragraph about silence after shelling carries weight because they were there. A parent writing about loss evokes empathy because they’ve felt it. AI can simulate those emotions, but it cannot feel them. Its version of “sadness” comes from statistics, not sorrow.

That difference matters. When you read a piece that moves you, what grips you is not just language — it’s recognition. You sense that another mind has wrestled with something real. Without that struggle, writing is empty mimicry. AI may produce coherent sentences, but it can’t produce meaning.

5. It Encourages Intellectual Laziness

There’s also a cultural cost to normalizing AI-generated content: we risk forgetting how to think deeply. Writing is hard for a reason. It forces you to clarify ideas, make choices, and confront what you don’t yet understand. Offloading that process to an algorithm might save time, but it also short-circuits reflection.

When students use AI to write essays or marketers use it to fill blogs, what’s really being lost is the practice of thought. The more we rely on machines to say things for us, the less we know what we actually mean.

6. Quantity Isn’t Quality

Defenders of AI often argue that it boosts productivity. But productivity is not the same as value. The internet doesn’t need more content; it needs better content. Yet AI makes it trivially easy to flood platforms with low-effort material — the kind of SEO sludge and “10 tips” articles that already clog the web. The more of it we produce, the harder it becomes for genuinely human voices to be heard through the noise.

The Bottom Line

AI writing tools are impressive — as assistants. They can help brainstorm, outline, or summarize. But when they become the source of the content, the results are soulless imitations of real communication. Writing is a human act: it’s how we wrestle with thought, emotion, and truth.

AI can mimic language, but it can’t inhabit it. And that’s why, for all its fluency and flash, most AI content remains what it is — technically competent, instantly forgettable, and, ultimately, crap.

A quick analysis of ChatGPT’s article

ChatGPT’s article covers “more” in the sense that it has six numbered sections. But what does it actually say?

1. “AI imitates, doesn’t think” – everyone knows this
2. “AI has no voice” – repeats point 1 with literary name-drops
3. “AI flattens creativity” – repeats point 1 again
4. “AI can’t replace lived experience” – repeats point 1 yet again
5. “It encourages laziness” – vague assertion
6. “Quantity isn’t quality” – another restatement

It’s the same observation dressed up six different ways. The kind of article you skim and forget immediately because you’ve read versions of it a hundred times.

Our bottom line (from us, not ChatGPT)

You can publish efficient content that nobody connects with, or you can publish content that achieves something.

Fancy the latter? Start a project: