Open five company blogs in your sector this morning and you will struggle to tell them apart. The same confident cadence, the same tidy sub-headings, the same reasonable-sounding paragraphs that commit to nothing. This is not because everyone got worse at writing. It is because everyone got the same machine at the same time, and the machine has a house style.
When a capability becomes universal, it stops being an advantage. The question is no longer whether you can produce content. Everyone can. The question is whether anything you produce could only have come from you.
Adoption is now near-total, so it differentiates no one
The scale of this is not speculative. Stanford's AI Index documents how quickly generative tools have moved from novelty to standard business infrastructure, with adoption climbing across functions and industries year on year (Stanford HAI AI Index Report). Gartner has tracked the same rush, with generative AI among the most rapidly adopted technologies enterprises have ever deployed (Gartner on generative AI). When a tool is that widespread, owning it proves nothing. It is table stakes, and table stakes do not win hands.
The predictable result is a flood of competent, forgettable output. Fluency has been commoditised. What has not been commoditised, and cannot be, is a specific human deciding that the consensus is wrong.
The moat is judgement, specificity, and a real position
A model is trained on the average of what already exists. By design, it regresses toward the middle. That makes it excellent at producing the expected sentence and structurally incapable of producing the surprising, earned, slightly risky one. Three things resist the averaging:
- Judgement: a call on what matters and what does not, made by someone accountable for being wrong.
- Specificity: the number, the client, the failure, the detail that only your experience contains.
- A real position: a stance you would defend in a room, not the balanced summary of every stance.
This is why the flattened, agreeable register that AI defaults to is such a liability. We have written about the death of corporate tone for a reason. The safe institutional voice was already forgettable. Now a machine can generate infinite quantities of it, for free, for everyone. Sounding like that is no longer neutral. It is a signal that no one is really home.
If a model could have written it without you, it will not distinguish you from anyone the model also wrote for.
AI answers reward the distinctive source, not the average one
There is a second, sharper reason to care. The systems deciding what gets surfaced are increasingly AI systems, and they are learning to prefer clear, attributable, specific sources over interchangeable content. As AI answers replace search results, the winners are the named voices with a defined position that a model can confidently cite. Generic content has nowhere to rank. It gets absorbed into a summary that credits no one.
So the paradox is real. The more everyone leans on AI to produce sameness, the more valuable it becomes to be the source that is unmistakably not average. Distinctiveness is not just a brand preference now. It is what makes you legible to the machines doing the recommending.
Use the tool, keep the point of view
None of this is an argument against AI. Use it to draft, to research, to get past the blank page, to handle the mechanical parts of the work. The error is letting it supply the thinking. The division of labour that holds up is simple: the machine can help you say it, but the position, the specifics, and the willingness to be wrong in public have to be yours. That is the part no competitor can copy by buying the same subscription, because it is not a subscription. It is a person.
Ripple helps leaders build a voice that reads as unmistakably theirs, the kind of point of view no shared tool can reproduce. If you are worried your content is starting to sound like everyone else's, we are happy to help you sound like you again.




