The editorial jobs are not disappearing. They are moving upstream.

For a while I treated each job advert as an anecdote. Then there were enough of them that they stopped looking like anecdotes.

Mintago’s Content Lead is expected to own the brand voice and build the content engine end to end, connect the same argument to both HR and finance buyers, work with sales and customer success, and use AI for ideation, research and scale while keeping the voice and judgement human. The advert is unusually explicit about where the company believes the human value still sits.

Nscale’s Senior Content Editor is asked to define the company’s editorial vision and narrative architecture, align it with positioning and commercial priorities, build repeatable programmes, and contribute to long-term content infrastructure, governance and knowledge management. It is an editor role whose own description says the person should build systems, not just assets.

Paddle is hiring a Head of Content to lead editorial direction and narrative consistency across long-form thought leadership, flagship reports, executive messaging and customer storytelling. The role also owns the standards, systems and governance that make the work coherent across the business.

Deepgram’s Head of Editorial Content is expected to mine stories across research, engineering, product, customers and leadership, then turn them into recurring editorial themes and formats that reinforce the company’s narrative over time.

And Noah Greenberg, the CEO of Stacker, regularly points journalists towards editorial jobs inside organisations they might never think to look at. His searches turn up titles such as VP Content, Editorial Director, Managing Editor, Head of Content and Data Journalist. He says he finds more than ten open editorial roles in a typical week this way.

I don’t think this proves writing jobs are safe, and it suggests something more interesting than that anyway. The claim isn’t that AI won’t replace writers. It’s that AI is changing where the editorial value sits.

Editorial work is moving upstream.

---

AI can produce the words. That is not the whole job.

The obvious AI story has been about production. Can it write the article, produce the press release, turn an interview into a customer story, generate social posts from a report? Increasingly yes, and that matters — some production work will disappear, some teams will get smaller, and roles built mainly around turning an agreed brief into competent copy will come under pressure.

But look at what these newer jobs actually ask a person to do. Decide what matters and what doesn’t. Translate complexity without losing the substance. Find customer evidence and shape a point of view from it. Advise executives. Build a narrative that holds together across press, social, campaigns, events and customer marketing. Watch the market closely enough to recognise which moment deserves attention. Listen to the response and feed what it tells you back into the organisation.

None of that is a writing task in the sense the debate usually means. They are editorial judgement tasks, and the writing is what happens after they have been done.

Maybe “editor” is becoming a bigger job, not a smaller one

For years content teams organised themselves around outputs. We need an article. We need a report. We need six customer stories. We need a campaign landing page. That made the writer visible, because the final asset was the visible thing — while a lot of the harder work stayed hidden inside the process: investigating, questioning, choosing evidence, rejecting weak claims, finding the argument, understanding the audience and deciding what should not be said at all.

AI is separating those two things, because it is now much easier to produce the visible output without having solved the invisible problem underneath it. That changes where the value sits. If an organisation can produce a hundred reasonable articles, the question stops being whether it can make content and becomes whether any of those articles deserved to exist.

This is where Editorial Intelligence comes in for me

I’ve been developing something I call Editorial Intelligence, which I initially thought of as a better way to connect research, narrative and content. The more I work on it, the less convinced I am that “content” is the useful centre of gravity at all.

The real problem is organisational knowledge. Companies know enormous amounts, and the knowledge is scattered across research, customer calls, sales conversations, product teams, events, executives, support tickets, market data and meetings. The work is establishing what matters, what is missing, what can actually be proved, and what coherent narrative can be built from what survives that. Then — and only then — you have a production problem.

My working model increasingly looks like this:

signals → evidence → judgement → narrative → AI-assisted production → activation → response → new evidence

That’s much closer to what I’m seeing in these emerging roles than the old model of receiving a content brief and producing an asset.

The pattern is appearing across different kinds of company

One reason I find these examples useful is that they are not all the same kind of company. They range from financial wellbeing and payments to AI infrastructure and voice AI. A handful of live job adverts still cannot tell us what will happen to the whole labour market, but the variation makes the pattern harder to dismiss as a quirk of one company or sector.

Across them, the role sits close to business strategy and organisational knowledge. Mintago combines brand voice with thought leadership, buyer understanding, sales partnership and AI-assisted execution. Nscale joins narrative architecture to commercial priorities, governance and knowledge management. Paddle connects editorial direction to executive messaging, customer storytelling and operational systems. Deepgram wants its editor to mine research, engineering, product and customer evidence.

None of these organisations is hiring only for someone who can make words. They are hiring people to decide what evidence matters, shape a point of view, create coherence across functions and build the system that lets the story travel.

The titles differ from one company to the next. The underlying capability being described looks increasingly like the same job.

The career path may be changing too

This matters to experienced writers and editors because the traditional career ladder has often been surprisingly narrow.

Write.

Become a senior writer.

Edit other writers.

Manage a content team.

Move into management.

But what if management isn’t the only way “up”?

AI may create more space for a different kind of senior individual contributor — someone who combines editorial judgement, research, narrative, customer understanding, AI workflows and organisational knowledge, and who can move between evidence and execution without being defined by either.

That’s increasingly how I think about my own work, too. I still call myself a writer — fifteen years of writing and journalism is a large part of why I can exercise this kind of editorial judgement at all. But “content writer” increasingly describes only the production layer of what I do. I’m more interested in the layer above it: evidence, editorial judgement, narrative and the systems that connect them.

That role doesn’t yet have a settled title. Depending on the organisation it sits somewhere inside editorial strategy, thought leadership, content strategy, communications, narrative, executive positioning or customer storytelling — and the absence of a name may itself be part of the signal, because the work appears to be reorganising faster than the vocabulary for it.

I don’t think the future is “human writers versus AI”

That framing feels increasingly unhelpful. The stronger question is where human judgement belongs in a system where production has become abundant, and my answer, at least for now, is upstream.

Closer to the evidence.

Closer to the customer.

Closer to the argument.

Closer to the decisions about what an organisation should say, why it can legitimately say it, and what it should learn from the response.

The more of these roles I see, the less Editorial Intelligence feels like an attempt to defend an old profession, and the more it feels like an attempt to describe where part of that profession is going next.