Most organizations have started using AI at the most obvious place: production.
Writers use it to draft articles. Social media managers use it to create posts. Designers use it to develop visuals. And it works. AI can make all of those things faster.
But giving everyone access to AI doesn't automatically make the communication function better. In fact, as it becomes easier for everyone to produce more, another part of the job becomes much more important: deciding what should be produced in the first place.
What is actually worth saying? Which stories matter to the business? What sounds like us? What is good enough to publish? And who takes responsibility for those decisions?
These are editorial questions. And as AI takes on more of the production work, communication teams need to spend much more of their time answering them.
Everyone Got AI. Now What?
AI adoption in marketing has moved quickly. Jasper's 2026 State of AI in Marketing, based on a survey of 1,400 marketers, found that 91% now actively use AI in their work. Half said AI is helping them bring work to market faster.
That's a significant productivity gain, but it also creates a new problem.
When every writer, social media manager and marketer can produce a first draft in seconds, producing the first draft is no longer the difficult part. The challenge moves towards choosing the right idea, finding the strongest angle and making sure the finished content is genuinely useful, accurate and recognizably yours.
Without that layer, AI can simply make inconsistency easier to scale. One person prompts for a blog post, another for a LinkedIn post and someone else for a video script, each with their own instructions and their own idea of what good looks like.
Research from IAB gives some indication of the risk. In a 2025 survey of 125 U.S. advertising executives involved in AI use, more than 70% reported at least one AI-related incident, including inaccurate, biased or off-brand outputs. Still, fewer than 35% planned to increase investment in AI governance or brand-integrity oversight over the following year.
The answer isn't necessarily more rules or another layer of approvals. But somebody still needs to set the standard.
Editorial Judgment Becomes More Valuable, Not Less
This is where the idea of the editor-in-chief becomes useful, even if marketing teams don't need someone with that title.
In a newsroom, the editor-in-chief doesn't write every article. Their job is to understand the audience, decide what deserves attention, maintain a point of view and make sure what gets published meets the standard of the publication.
Marketing and communication teams increasingly need that same function.
Someone needs to understand the business well enough to know which stories matter. Someone needs to recognize when an idea is interesting and when it is just more content. Someone needs to question vague claims, protect the brand's point of view and decide when human expertise is essential.
This doesn't have to be one person. Depending on the organization, those responsibilities might sit across marketing leadership, communications, brand and subject-matter experts.
What matters is that those responsibilities exist — and that people actually have time to take them seriously.
That's one of the more interesting opportunities AI creates. If less human time needs to go into first drafts, adaptations and production, more of it can go into asking better questions, finding stronger angles, understanding what customers care about and deciding what the company genuinely has something interesting to say about.
In other words, AI doesn't just make editorial judgment more important. It can give communication teams more time to exercise it.
Give AI Clearer Boundaries
More editorial responsibility also means being clearer about what AI should and shouldn't do.
Not every piece of communication carries the same risk. Using AI to explore ten possible headlines is very different from asking it to make a financial claim, respond to a crisis or formulate a company position on a sensitive issue.
The same goes for brand voice. Telling AI to be “warm but professional” leaves a lot open to interpretation. If AI is going to play a larger role in production, teams need a clearer understanding of how their brand actually communicates: what it does, what it avoids and what a good piece of content looks like.
Human review matters here too. IAB's 2026 State of Data research found that mandatory human oversight is among the most common measures marketers are using or planning to use to address AI challenges, although fewer than 40% reported having or planning any individual solution included in the study.
The point isn't to build an enormous governance structure around every LinkedIn post. It's to make deliberate choices about where AI can work independently and where people need to be involved.
Someone Still Has to Own the Result
There is another reason editorial oversight matters: when something goes wrong, your audience doesn't care which part of the workflow caused it.
An inaccurate claim is still your inaccurate claim. An off-brand post still came from your company. If AI produces something inappropriate and nobody catches it before publication, the fact that AI was involved doesn't make the organization any less responsible for it.
“The challenge isn't whether to use AI,” says Peder Bonnier, CEO at Storykit. “It's whether you've built the editorial intelligence to make AI worth using. Without that layer, you're just producing more noise, faster.”
As AI takes on more of the production, human responsibility doesn't disappear with it. Someone still needs to make sure what goes out is accurate, appropriate and good enough to carry the company's name.
AI Can Produce. Humans Still Have to Choose.
Making production dramatically easier also means you can suddenly make content from almost anything: A report can become social posts. An interview can become an article or a video. A presentation can provide ideas for weeks of communication. AI makes it much easier to work with the knowledge and expertise an organization already has and turn it into different formats.
And producing more can absolutely be a good thing. More useful content means more opportunities to reach people, build familiarity and stay present. If you have ten worthwhile things to say, being able to publish all ten is an advantage.
The important word there is worthwhile.
Someone still has to find the useful ideas inside the source material and decide which ones deserve attention. Making production easier doesn't make those choices less important. If anything, it makes them more important because there is no longer the same practical limit on how much you can produce.
“Storykit automates up to 95% of the production process, so we can apply the human touch where it matters most – the message and the strategy,” says Fredrik Strömberg, Chief Innovation Officer at Storykit.
What Does This Mean for the Communication Team?
You probably don't need to hire an editor-in-chief.
But it is worth asking whether anyone in your current setup is clearly responsible for the work an editor-in-chief would traditionally have done.
Who decides what is worth talking about? Who knows what good looks like? Who makes sure the company's communication has a recognizable point of view rather than becoming a collection of independently generated outputs? Who has the authority to say that something isn't good enough to publish?
As AI takes on more production, those questions become harder to avoid.
“Every organization has more to say than they realize,” says Jonna Ekman, Marketing Director at Storykit. “And the problem isn't creating content anymore. It's having the editorial clarity to know what's worth saying, and the system to say it consistently across channels, markets, and formats.”
Start With One Workflow
Don’t try to redesign your entire communication function overnight. Start with one type of content your team already uses AI to produce and look at how the work actually gets done.
Who decides what the content should be about? What does AI do, and what still requires human judgment? Is everyone working towards the same idea of what good looks like? And, most importantly, who is responsible for the final result?
Then look at where your team is still spending time on work that AI could reasonably take on, and where more human attention would actually improve the outcome.
The goal isn't to add more process. It's to make the division of work more deliberate: let AI handle more of the production, and make sure people have the time and responsibility to handle the judgment.
The marketing teams that make the most of AI won't simply be the ones that learn how to prompt it best. They'll be the ones that get clearer about what deserves to be said, what good looks like and where human judgment matters.
