For generations, the early stages of ideation have been treated as a kind of mysterious interior weather system—a landscape where intuition and uncertainty coexist, where fragments gather without form, where the smallest spark can become a structure and the clearest structure can suddenly unravel. Writers, filmmakers, choreographers and architects have described this space with a shared reverence. Octavia Butler wrote about the “unfocused hour” when connection begins to form. David Adjaye spoke of ideas arriving from “disparate signals that have not yet settled.” Marina Abramović described her early creative process as “waiting for the moment where the idea chooses you, not the other way around.” These expressions capture a truth that technology has never been able to replicate: the first stage of an idea is emotional before it becomes logical, intuitive before it becomes communicable.

The rise of generative AI has challenged and expanded this landscape. Tools like GPT-5, Midjourney v7, Claude 3.5 and Runway Gen-3 do not wait for inspiration; they produce material whether or not the creator has found their intention. They can reveal possibilities that would have taken days to explore manually. They can assemble surprising connections, shift metaphors, generate visual rhythms and articulate conceptual options with the kind of fluency that once required entire teams. To many creators, these tools feel like an extension of the mind—an external imagination capable of responding instantly. And yet, the heart of ideation remains stubbornly human.

This is because generative systems, no matter how impressive, do not originate ideas in the philosophical sense. They remix. They recombine. They reconstruct. Their creativity is emergent, not intentional. They work by interpreting patterns—patterns shaped by culture, by the internet, by collective histories—and offering statistically coherent reflections of them. A model may surprise you, but it does not surprise itself. It does not experience the delightful vertigo that accompanies a truly new thought. It does not feel the gravity of an idea that matters. And without that gravity, it cannot steer.

At Copy Lab, we see this distinction not as a limitation of AI but as the foundation of the sacred partnership between humans and generative systems. AI expands the field of exploration; humans determine which pathways carry significance. The model can open doors, but it cannot tell you which one leads toward the story you need to tell, the tension you want to expose, or the emotion you want to evoke. It can offer breadth, but it cannot provide the compass.

The strongest ideas in history were rarely linear. They emerged from contradiction. Consider how James Baldwin shaped his essays through an ongoing dialogue with the world—never arriving at clarity without passing through complexity. Or how Sofia Coppola crafts films by lingering on emotional atmospheres rather than clean narrative arcs. Or how Björk works through layers of sound until she finds the one that feels “alive.” Their processes were not built on producing endless drafts for efficiency; they were built on listening—to memory, to instinct, to the internal pulse that signals when an idea is beginning to take form.

GenAI accelerates exploration, but it does not accelerate this listening.

If anything, it demands more of it. When the tools can generate dozens of reasonably strong directions, creators must sharpen their ability to sense which direction carries the potential for depth. When the systems deliver polished surfaces, humans must look beneath those surfaces for meaning. When the interface produces instant clarity, the creator must stay connected to the uncomfortable ambiguity from which real ideas grow.

Ideation in the age of GenAI is therefore both easier and harder. Easier because the pathways are abundant. Harder because abundance requires commitment to the one path that matters. This is the paradox that modern creators face: the tools can do the wandering, but only the human can discover where they are actually trying to go.

For Copy Lab, this is not a challenge—it is the opportunity we are built for. GenAI allows teams to expand their conceptual reach, to test ideas rapidly, to stretch into new visual or narrative territories without the constraints that once limited experimentation. But the purpose of this expansion is not to reduce ideation to a mechanical process. It is to give creators a richer landscape from which to draw their interpretation, their vision, their instinct.

In other words: AI widens the horizon, but humans choose the direction.

The future of ideation will not be defined by how quickly we can generate ideas, but by how deeply we can interpret them. The role of the creator becomes more essential, not less. And the sacred partnership between human intuition and generative intelligence becomes not a compromise, but the most powerful creative model our era has yet seen.

/Carl-Axel Wahlström, Creative Director Copy Lab, 2025