Here is the thing nobody in the AI-and-content conversation seems willing to say:
AI, in a content operation, is not for writing. It’s for thinking.
I run content for several clients across fitness, finance, and corporate strategy. I also write a significant volume of the content myself. I use AI every day.
And in three years of integrating these tools into actual client workflows, the use cases that made the most difference reliably have nothing to do with generating text.
What we mean by editorial advisor
An editorial advisor is someone who reads what you’ve produced, asks questions about it, and pushes back. “Is this actually the argument you want to make? What’s the evidence for this claim? Have you considered the objection in the third paragraph?”
It’s a function that, historically, you needed an editor for. A good editor is expensive and rare. AI is neither of those things.
AI works as an editorial advisor in 3 main stages of the content process.
i. During research
Before I write anything substantial, I use AI to take the opposing position. If I’m writing a piece about why content audits matter, I ask the model for the strongest possible argument against content audits. Not to use that argument, but to be sure I’ve addressed it. This discipline alone has made my writing better.
II. During outlining:
I share an outline and ask the model to find the weakest section — the place where I’m asserting without supporting, or the logic has a gap. It’s not always right, because AI is often wrong about everything, but it’s right often enough that this has become my standard practice.
III. During revision
I paste in a draft and ask specific questions: “Does paragraph 4 follow from graf 3?” “Is the conclusion doing something different from the introduction?” I ask it to evaluate structure and logic so i know what to attack in the rewrite.
Implications for your content operation
If you’re a brand or marketing team, the implication is this:
AI is a multiplier on human & editorial judgment, not a replacement for it.
The teams getting outsized results from AI tools are the ones who had strong editorial standards in the first place: tools amplify what’s already there, always.
teams that adopt AI to substitute for editorial judgment & writerly quality end up with more content of the same or worse quality than they had before, which was: not enough.
The question isn’t “how much of this content can AI produce?” The question is “what does a human have to contribute to this content for it to be worth producing at all?”
Answer that one first. Then figure out where AI fits.
Working out where AI fits in your content operation is a strategy problem, not a tech problem. The operational side of that — how to actually structure a content operation that uses AI effectively — has its own guide.
Luckily, that’s exactly what I do.
Jacob Clifton is the principal of Clifton Creative, an editorial strategy consultancy based in Austin, Texas. He spent fourteen years as a flagship staff writer at Television Without Pity and has written for Tor.com, Vulture, BuzzFeed News, and the Austin Chronicle.