There’s a conversation happening in boardrooms and marketing teams across the world right now and it starts something like this: “We have AI. Can we spend less on human expertise?”
It is an understandable question. AI tools are proving invaluable – they can produce content, planning and ideas in seconds, but the question contains a fundamental flaw: understanding what marketing expertise (that is, ‘human’ work) actually is.
Consider AI the newest instrument to the marketing orchestra. Without a conductor guiding it, it can’t produce music, it just produces noise.
Here's the thing: AI accelerates, but it doesn't strategise
We agree that the productivity gains from AI in marketing can’t be denied. According to research from McKinsey, revenue increases from AI use are most commonly reported in marketing and sales, and high-performing organisations are using AI to redesign entire workflows, not just automate individual tasks. A separate industry report by CoSchedule found 88% of marketers now use AI in their daily roles, and 83% report measurable gains in productivity since adoption.
But here is the clincher: only 26% of those organisations have actually figured out how to generate tangible value from it. So, how to close the gap between adopting AI and deploying it? You guessed it – human expertise.
While AI can be used to execute actions within a defined system, it can’t determine what the system should be in the first place. It doesn’t understand your business market position, nor can it identify where your competitors are going wrong in that context. It doesn’t understand creative limitations within a brand guideline, or where it can test the boundaries. All that is covered by human ideation and strategy mapping.
In short: AI can scale your content and productivity, but it won’t scale your judgement.
The homogenisation problem
There is another glaring risk that proponents of AI-only marketing rarely mention: when every brand uses the same tools fed with the same prompts, trained on the same internet, the content starts to look and sound the same.
It’s evident even in these early days. Researchers studying generative AI across creative writing, advertising content, and idea generation have identified a consistent pattern of homogenisation — a convergence toward mainstream, predictable outputs that narrows rather than expands creativity. The mechanism is twofold: the AI itself amplifies patterns from standardised training data, and human users tend to gravitate toward and build upon AI suggestions, further narrowing down the landscape of thought and ideation.
This poses a big threat to a brand’s strongest weapon: voice. And audiences are more clued in on that homogenised content than you might assume. Consumers can identify AI-generated content with increasing accuracy, trusting it less. Brand consistency with a distinctive voice increases revenue by 10–33%, which means that ultimately, the brands that win will be the ones that stand out rather than keep up.
The role of marketing agencies
The question is not whether AI replaces expertise, it’s what expertise looks like in an AI-enabled environment.
Think of it this way: a spreadsheet did not replace the accountant. It eliminated hours of manual calculation so the accountant could spend more time on judgement, advice, and strategy that comes from years of expertise and contextual knowledge. AI is doing the same for marketing. With less time taken between concept and execution, experienced marketers can spend more of their capacity on the things only they can do: shaping brand narrative, identifying cultural moments to stay reactive on socials, optimising campaigns based on their years of industry experience and knowledge of your business, asking the uncomfortable questions about whether a campaign is actually right, and building the systems that ensure quality at scale.
McKinsey’s research on AI high performers finds that the organisations extracting the most value from AI are those with defined processes for when model outputs require human validation — not those using AI to replace human oversight entirely.¹ The expertise required to build those processes, interpret those outputs, and know when to override them is irreplaceable.
Final thought
The brands winning in 2026 will not be the ones that produced the most AI-generated content. They will be the ones that used AI’s speed in service of a clear, differentiated, human-led strategy.
Human marketers are able to understand cultural context and act reactively in ways that AI simply can’t replicate. As production becomes faster and potentially more homogenous, the human ability to create something genuinely distinct for your brand becomes rarer, and therefore more valuable.
