AI Will Not Take Your Marketing Job. It Already Took the One Below Yours.

There are so many conversations happening about AI and marketing jobs right now, and they rarely meet each other.

In one corner, the platforms, consultancies and the conference keynotes are telling you that AI augments marketers, it does not replace them, that human creativity is irreplaceable, that the future belongs to those who adapt, and so on. They are branding it AI-assisted marketing, a polished version of a story, told from top down, meant to sound reassuring if you don’t look too closely at who’s telling it.

Then there's the other conversation happening in team meetings, in ‘in-betweener’ Whatsapp groups, and over water-after-work; and in that conversation, marketers are terrified. Job postings are falling, entry-level roles are disappearing, senior leaders are being cut and nobody’s replacing them, and the headline that 65% of marketing jobs may not survive AI didn't come from some random flop blog, it’s from Adweek.

The truth sits somewhere between the institutional reassurance, which clearly has not passed on the memo to the recruitment team, and the existential dread, hence this piece.

The people selling you AI, well, have something to sell

The loudest voices arguing that AI creates more efficiencies for marketers than it eliminates jobs are, almost uniformly, the people selling the tools, running the platforms, or leading organisations whose CFOs are using AI as cover for cost-cutting guised in driving efficiency. That doesn’t make them wrong, but it should make you stop and ask who is not in the conversation?

And here's where it gets good, because the numbers don't fully support either side. Skills in AI-exposed roles are changing 66% faster than in the least exposed ones. Marketing postings fell sharply through 2025 and into 2026, and almost a quarter of organisations that cut senior marketers chose not to replace them. The roles going first are entry and mid-level ones: the analyst building the weekly performance deck, the executive writing a first draft, the coordinator managing the social media content calendar. A senior marketer with the right tools can do all of that before his bon appetit.

Respect is due to those who already moved fractional, they read the direction of travel early. Those still job hugging without adapting will find themselves left behind because the workscape changed around them while they were holding on to the old version of it.

The question one should be asking

Those entry-level roles produce senior marketers, this is the phase where the craft gets passed on. These are the roles where you learn to brief a creative, read a room with a client, optimise a campaign that underperformed, and develop the kind of judgment that no AI can replicate because it's built from painfully lived experiences.

If those roles disappear, who trains the next generation of marketing leaders?

And those of us who are hiring need to be more honest about what we are actually asking for. If you expect a marketer to lead a STP exercise, develop campaigns, write genuine copy, then launch and control the whole thing across channels, that’s not a job AI solves. AI can accelerate parts of the execution, but it can’t replace the strategic and creative judgement that makes the execution run properly in the first place.

A practical example: AI can tell you exactly where a lead drops off in a journey, show you the stage in the funnel where people lose interest, and it can do that faster and more precisely than any of us could. But put several campaign concepts in front of it and ask which one to run for a specific audience in a specific market at a specific moment, and it will give you a confident answer that could be completely wrong. It doesn’t have the instinct, the cultural reading, or the experience of having gotten it wrong before and learnt from it. That’s still a human call, and it is the call that determines whether the campaign works or doesn't.

I have now several times received overly confident answers from AI that were completely wrong, so conflating execution with strategy is how you end up with teams full of prompt engineers and no one who actually understands marketing.

The efficiency that's not showing up

There’s another part of this conversation that directly challenges the narrative that AI is making everything cheaper and faster.

In Feb 2026, a study surveyed 6,000 C-levels across the US, UK, Germany and Australia. Nine out of ten reported that AI had no measurable impact on productivity or employment at their organisation over the past three years. Deloitte found that, while 74% of organisations want AI to grow revenue, only 20% have actually seen that happen.

Scott Galloway made this point sharply on The Diary of a CEO. The AI CEOs need to justify their valuations, and the way they do that is by promising massive savings that will flow to their customers' bottom lines. One of two things has to happen in the next few years: either those valuations get cut dramatically, or there has to be large-scale labour destruction to deliver promised savings. The catastrophising, as he put it, tends to peak right around the time someone is cashing out.

For marketers, the takeaway is practical, don’t assume that adopting AI automatically means your company is saving money or becoming more productive. In many cases, the cost of implementation, the time spent learning the tools, the integration headache, the culture shift that’s required and the volume of output that now needs human review is only adding cost.

AI can make an individual marketer more efficient at specific tasks, but scaling that across an organisation is a completely different proposition, and most companies are not there yet.

Which brings me back to the question I raised earlier: if the efficiency gains are not materialising at scale, and the junior roles are being cut anyway, then isn't the profession cutting into the muscle that produces its future leaders?

On tools, overwhelm, and the lie of the stack

To the marketers who are struggling to keep up with all of this: you are not behind, they are overwhelming us by design.

The martech landscape now has 15,000+ solutions, and organisations use a barely a third of the capabilities they are paying for; we've been sold fragmentation as sophistication, and complexity as progress.

The landscape is shifting in two directions at once, legacy tools are exiting at the fastest rate in years, and AI-native tools are entering faster than the old ones leave. Platforms like HubSpot, Adobe and Salesforce are embedding AI directly into their workflows, becoming orchestration layers around which everything else connects. The marketer whose expertise was built on mastering a specific stack of tools will need to think differently, because the skill that holds value is the ability to know when the input is inaccurate, when the output is off, and to make the decisions that the platform can't make for you.

Part 2: AI took the job below mine. Here's what I'm doing about it.