Using AI To Craft Your Marketing Content Is Damaging Your Brand

I've spent the past few months properly testing ChatGPT inside real marketing work. Not casually playing around with it. Actually trying to integrate it into long-form blog writing, LinkedIn content, positioning work, messaging, SEO pages, strategic thinking, and commercially-sensitive copy.

I chose ChatGPT, because it's currently what most people - including potential clients of mind - use.

I followed all the best practice advice. Gave detailed context. Provided examples of my writing. Explained my audience. Corrected outputs repeatedly. Explained which phrasing patterns I hate. Explained which rhetorical structures scream AI. Identified weak thinking. Gave examples of stronger writing. Did everything you're supposed to do.

The output still failed in ways that matter commercially.

Not occasionally. Repeatedly.

And I'm not talking about obviously-bad content that anyone would spot. I'm talking about content that looks fine at first glance but slowly flattens everything that makes a business commercially-distinctive.

What I Actually Did Properly

Before anyone jumps to "you're just bad at prompting" - I wasn't.

I gave ChatGPT:

• Detailed context about my business, positioning, and messaging

• Clear explanation of my target audience

• Multiple examples of my writing style

• Explicit instructions about what to avoid

• Repeated corrections when outputs were weak

• Explanations of which themes I'd already covered

• Examples of commercially-grounded writing vs. AI sludge

• Specific criticism of paragraph flow, formatting, tone, cadence

• Clear identification of fake depth and consultant-speak

I spent hours on this. I approached it properly. I wasn't being lazy or vague. I even set up Custom GPTs with very detailed and specific prompts.

That matters, because most people dismiss AI failures as bad prompting. This wasn't that.

The Problems That ChatGPT Kept Repeating

ChatGPT kept reverting to the same patterns, even after I'd explicitly corrected them multiple times.

Generic business language. Repetitive phrasing. Tidy rhetorical contrasts. "It's not X, it's Y" structures everywhere (which are fine by the way - just not all the time). Over-explaining obvious points.

Lots of the content produced sounded like every other piece of so-called LinkedIn thought leadership. Over-structured cadence. Commentary that sounds intelligent without actually saying much. One-line paragraphs stacked for dramatic effect.

Like this.

Exactly. Like. This.

It drfited back into themes I'd already covered. Rewrote entire sections instead of editing what was there when asked. It made every article start to sound interchangeable.

Here's what really struck me though: ChatGPT often understood the criticism afterwards. I'd point out why something failed, and it would explain back to me exactly why the writing was weak. Then, in the next output, it would repeat many of the same mistakes.

ChatGPT knows what good writing looks like. It can analyse what went wrong. But it can't consistently apply that judgment across multiple outputs.

Why This Matters Commercially

The danger for businesses and marketing teams isn't necessarily producing AI-created content that everyone spots immediately. As long as it's doing the job, then I think that's ok.

For me, the danger is producing content that looks fine but slowly strips away what makes your business distinctive.

Flattening differentiation. Weakening positioning over time. Making your brand sound like everyone else. Increasing content volume while reducing originality. Slowly removing the sharp edges that make someone choose you over a competitor.

All this matters enormously in B2B. In founder-led businesses. In agencies. In specialist firms competing on expertise and positioning.

Your positioning isn't just what you say you do. It's how you say it. The things you state. The things you choose not to say. The way you frame problems. The specificity of your examples. The credibility that comes from lived experience, not internet phrasing.

AI defaults towards statistically-common patterns. That's literally how it works. Which means it defaults towards making you sound like everyone else.

And that's not great.

AI should serve a strategy. But if the underlying strategy is muddled and inconsistent, so too will the results be.

AI Struggles With Maintaining Distinctiveness

I asked ChatGPT as a test to write several blog articles on different marketing topics.

Each one started sounding more and more like the others that came before.

Same cadence. Same rhetorical structures. Same progression from problem to insight to solution. Same consultant-narrator voice.

If I'd published all of them, someone reading my site would think I'd lost the ability to vary my writing. Or that I'd outsourced everything to the same ghostwriter who only knows one template.

That's commercially dangerous. Readers notice patterns. They notice when everything starts sounding the same. They notice when the distinctiveness drains out.

And in B2B marketing, where expertise and credibility matter, sounding generic is death. You're not competing on volume. You're competing on "does this person actually know what they're talking about?"

What AI Is Actually Good For

This article isn't saying "AI is useless." AI is genuinely useful for idea development, research, workflow acceleration, summarising, structural support, challenging thinking, and speeding up admin tasks.

I've used it effectively for all of those things.

But positioning? Messaging? Thought leadership? Brand distinctiveness? Commercial credibility?

Those still need strong editorial judgment. Human judgment. From someone who notices when writing is drifting into generic territory. Someone who can feel when a sentence sounds like it was optimised for LinkedIn engagement rather than actual commercial persuasion. Someone who knows the difference between content that looks professional and content that actually positions the business well.

Everyone's Recommending AI - And That's A Problem

Here's what worries me the most though.

A lot of agencies, consultants, and internal marketing teams are being told to integrate AI into their workflows. Some are recommending it to clients. Some are rolling it out across entire teams without properly pressure-testing it first.

But when it comes to producing content, most businesses haven't asked the right question.

It's not a question of whether AI can help produce content. It can. The question is whether it can consistently protect what makes your business commercially distinctive?

That's a very different question. And, based on my test, it can't - or at least ChatGPT can't. I'll be testing others over the coming months to compare.

Most marketing teams have probably tested whether AI can write a blog post. Whether it can produce LinkedIn content. Whether it can draft an email. They haven't tested whether it can do that repeatedly over time without slowly flattening their positioning into generic mush.

And by the time they notice that it has, they've published six months of increasingly interchangeable content that's made their brand less distinctive and buyers less likely to commit.

What Proper AI Testing Actually Looks Like

Proper testing isn't writing three blog posts and thinking "this works." Proper testing is:

• Using it across multiple related topics and seeing if distinctiveness holds

• Checking if outputs start sounding interchangeable

• Watching for drift back into generic business language

• Seeing if it can maintain a specific voice under commercial pressure

• Testing whether it can handle nuance without over-explaining

• Checking if it defaults to common phrasing patterns

• Seeing if your audience would notice the difference

Most businesses aren't doing that. They're doing surface-level testing and assuming that's enough.

Worse still, they're just rolling AI out and believing that everything must be perfect - "because AI produced it".

What This Means For B2B Businesses

If you're a founder, CEO, or senior marketing leader, you should be asking harder questions before rolling AI out across your marketing.

Because if your marketing starts sounding like everyone else's, you're not saving time. You're eroding your positioning while producing more content.

And in B2B, where differentiation is already hard and buying decisions are heavily influenced by perceived expertise and credibility, that's commercially expensive.

What This Means For Agencies

If you're an agency recommending AI tools to clients, then it's a similar story, and you should be pressure-testing them properly first.

Because if you roll out AI tools that slowly flatten your clients' brands into generic sameness, you're not helping them. You're making them less competitive while charging them for the privilege.

And when they eventually notice their marketing sounds like everyone else's, they're going to blame you and you'll lose the contract.

The Strategic Judgment Gap with AI

AI can produce content. But it can't consistently apply strategic judgment.

It can write a blog post. It can't tell you if that blog post strengthens or weakens your positioning relative to competitors.

It can draft messaging. It can't tell you if that messaging will resonate with buyers who are choosing between you and three other similar businesses.

It can generate ideas. It can't tell you which ideas are commercially sharp and which ones are just statistically common business phrasing that sounds intelligent without actually differentiating you.

That judgment gap matters. And it's not fixed by better prompting. It's getting filled by experienced marketing leadership who can feel when something's drifting generic and pull it back.

So, Do You Take A Punt On AI?

AI tools are going to be part of marketing workflows. That's obvious. But businesses need to pressure-test them properly before assuming they're safe to roll out widely, especially when it comes to content creation.

Most businesses haven't done that yet. They're assuming AI limitations are prompting problems or that volume matters more than distinctiveness, or - worse still - that everything produced by AI is fine.

They're wrong.

And by the time they realise they've flattened their positioning into generic sameness, they've lost months of opportunity to actually differentiate themselves in the market.

If you're responsible for marketing and you're thinking about integrating AI more deeply into your workflows, especially for writing copy, test it properly first. Not three blog posts. Thirty. Not one LinkedIn post. Fifty. See if distinctiveness holds. See if your audience would notice the difference.

And if you're a founder or CEO watching your marketing team integrate AI without that level of scrutiny, you should probably ask harder questions about what's being protected and what's being eroded.

Because positioning matters. Differentiation matters. Sounding like yourself matters.

And tools that slowly strip those things away while producing more content aren't actually helping you compete.

If your marketing needs strategic oversight that protects what makes your business commercially distinctive - not just more content volume - let's talk. I work with B2B companies and agencies to make sure marketing actually strengthens positioning rather than slowly flattening it into generic sameness.