You’re reading a competitor’s latest blog post and it’s sounding very familiar.
Same structure. Same cadence. Same opening line about how “in today’s fast-paced world x,y,z”.
Jump over to LinkedIn and it’s looking like the equivalent of the village in Royston Vasey, a weird uncanny valley where every piece of content reads like the back of a cereal box. You read it and you know it’s just… off.
The problem is everything is the same and it’s quietly becoming one of the biggest threats to brand differentiation in the age of AI-generated content.
Default settings are running the show
Let’s get it straight, only you can sound like you. Everyone knows it, but the LLM’s are getting better at mimicking that spontaneity of language.
Large language models are trained on copious amounts of internet text, but they’re looking for the most common patterns, the most repeated phrases, the most “acceptable” sentence structures. When you ask an AI to write content without firm guardrails, it doesn’t reach for your brand’s unique personality. It reaches for the most middle of the road it can.
What you end up with is a kind of beige professionalism. Yeah it’s competent, readable but it’s utterly forgettable.
Suddenly every brand has a “passionate team committed to delivering exceptional results.” (because who wants to shout about an uninterested team delivering mediocre results?). Do you feel called out if every one of your newsletters opens with a rhetorical question and closes with a call to action that uses the word transform? You should be!
When everyone’s content is generated by the same underlying models with the same generic prompts (“Claude, write me…”), differentiation doesn’t just weaken it, it evaporates altogether.
The fix isn’t abandoning AI, it’s training it on you
Now we’re not saying that AI should be packed in, it’s a powerful tool that can really help us navigate some of the challenges with good blog writing. We’re saying with the right prompting it can support your content efforts without flooding the feed with beige blergh.
The same problem is entirely solvable. The issue was never AI itself, it was that AI is operating in a bit of a vacuum, without your brand’s identity embedded into every output.
So here’s how to close that gap:
1. Build a brand voice document that actually works
Imagine briefing a human copywriter that your brand voice guidelines were: “We’re friendly but professional”
“We’re honest and clear”
They’d certainly come back with questions! It’s too vague to be useful to a human so it’s definitely not useful for AI.
A strong brand voice document for AI use needs:
- Vocab lists: words and phrases you own, and words you actively avoid. What are the key things you want to say and how to say them.
- Sentence rhythm examples: short and punchy? Long and layered? Both, in specific contexts?
- Tonal spectrum: how does your voice shift between a complaint response, a product launch, and a thought leadership piece? How do you get the tone right for the piece?
- “We sound like this / we don’t sound like this” comparisons: side-by-side examples are worth a thousand adjectives, it gives the AI more to go on than ‘we sound upbeat’
We know that generic prompts produce generic output. So think about the way that you brief an AI. Treat them as if they are a talented writer who knows nothing about your brand on their first day.
Instead of: “Write a LinkedIn post about our new product feature.”
Try: “Write a LinkedIn post about our new product feature in the voice of [Brand]. Our tone is [specific description]. We avoid corporate jargon and never use phrases like ‘game-changer’ or ‘seamless experience.’ Our audience are senior operations managers who are time -poor and sceptical of hype. Lead with the problem, not the feature.”
The specificity isn’t just for fun, it actively suppresses the model’s tendency to try and retreat to its generic defaults.
Reference content: Any of your best performing, most on brand pieces that the model can pattern match against? Load them in!
When you’re working on prompting the AI, feed this in every time as your foundation.
2. Establish AI as the drafter, not the publisher
This is one of the most important things to keep in mind.
Treat AI output as a first draft that always passes through a human voice filter before it’s published. Your job isn’t to fix AI’s grammar, it’s to reinject personality and make sure it’s sounding like your brand.
This might mean:
- Replacing AI’s way of phrasing with words your brand actually uses
- Cutting the usual concluding paragraph that just repeats everything that was just said
- Adding a specific cultural reference or a piece of industry humour take that your audience would expect from you
- Adjusting the sentence length or use of second person to match your established voice (e.g us, we).
Think of it less as editing and more as translation from the generic to distinctly yours.
But…it won’t always be like this
AI will adapt quickly, the floor of content quality is rising and brands will stand out even more than before because the contrast between them and the beige middle has never been sharper.
The brands that will win the next phase of content marketing aren’t the ones who produce the most AI content. They’ll be the ones who use AI to maintain volume while doubling down on the distinctiveness that no model can generate without deliberate, sustained creative direction.
The Bottom Line
AI doesn’t have to make your brand sound like everyone else’s. But it will if you let it.
The solution for the future isn’t to struggle between creating more human content or less AI, it’s creating a better integration of the two.
So what are the main things to remember?
- Treat your brand voice as infrastructure – like a schematic for your content not decoration
- Make sure you build it into your AI workflows from the ground up
- Keep a human in the loop at every stage
In summary, the goal of using AI in content isn’t to replace creative thinking, if it’s left to its own devices all we’ll have is a sea of beige which no one reads. AI has its place in content creation as long as it’s fed the right ingredients, the human voice is always priority at the end of the day, who knows your brand better than you do?
Want to make your brand voice stand out?
Get in touch with the Shapes team at shapes.team
