Why AI didn't kill creativity

Two years ago the creative industry was in a panic. AI was going to replace designers, writers, editors and strategists. Some people predicted the end of creative work altogether, others celebrated that creativity had finally been democratised.

Since then I've integrated AI into more than 100 campaigns, for brands like Philip Morris and Bacardi. The reality turned out far less dramatic and far more interesting. AI didn't kill creativity. It killed busy work.

What most of us got wrong in 2023

The debate was framed badly from the start. It was never really about AI versus humans. It was about how humans use AI.

Most discussions focused on whether AI could create content, and that question was answered quickly. Of course it can. It writes, designs, edits and generates variations faster than any person. What almost nobody asked was whether AI understands why a particular creative choice works in a specific business context. That's where the real divide appeared.

What AI is genuinely good at

After two years of real-world use, the strengths are clear. AI is outstanding at production and iteration:

• producing large volumes of content quickly
• creating variations within clear parameters
• formatting and adapting content across channels
• handling repetitive execution without getting tired or bored

What it doesn't do well is decide what matters. That pattern isn't only my experience. In a Harvard Business School and BCG study of consultants using GPT-4, people finished tasks about 25% faster when the work sat inside what the researchers called AI's "jagged frontier". On a task outside it, the consultants using AI were 19% less likely to reach the right answer than those working without it. Brilliant at some things, confidently wrong at others.

Where people still matter most

Strategic creative work didn't disappear. It became more important. People still outperform AI at understanding business objectives and constraints, reading between the lines of what a client actually needs, making decisions about positioning and message, judging whether something will land emotionally, and building the trust and long-term direction a brand needs.

AI doesn't understand context unless you give it explicitly. It doesn't feel consequences. It doesn't know when something is technically correct but strategically wrong. That judgment still belongs to people.

Why some brands struggled and others didn't

The brands that struggled made the same mistake. They used AI for strategic thinking instead of strategic execution, and expected it to tell them what to say, who to target and why it would work. The MIT research Fortune reported on in 2025, which found that around 95% of corporate generative AI pilots delivered little or no measurable impact, points in the same direction: the problem is usually how the tool is used, not the tool.

The brands that did well treated AI like a very capable junior team member. Fast, productive and tireless, but dependent on clear direction. You wouldn't ask a junior to define your business strategy. You would absolutely ask them to help scale one once it's clear. That's exactly where AI performs best. The Content Marketing Institute's 2025 B2B research saw something similar: the most successful content teams stood out for having clear AI guidelines and building it into their daily workflows.

The unexpected result

Instead of replacing strategic creative directors, AI made them more valuable. Someone still has to set the direction the tools follow, judge the output on business impact rather than surface polish, protect the brand's voice at scale, and decide which ideas deserve to be made at all.

AI amplified execution. It didn't replace thinking. Teams that already had strong direction moved faster and made better work. Teams without it simply produced more noise, faster.

For creative professionals, the line didn't fall between people and machines. It fell between strategic creatives and task executors. Those whose value was tied mainly to repetitive production felt the pressure. Those who could connect creativity to business outcomes, and direct the systems instead of competing with them, became more important. That shift was coming anyway. AI just sped it up.

The better question to ask

If you're still asking whether AI will replace creativity, you're asking the wrong question. The useful one is this: how can AI amplify my creative direction instead of replacing it?

That's where the opportunity is. If you're curious which tools earned a place in my own work, I wrote about that in the tools I use in 2026.

And if you're figuring out where AI belongs in your team, I'm happy to think it through with you. You can read more about how I work, see selected work, or book a 1-1.

— Joost