If you have been part of any conversation about AI and creativity, you have heard some version of this: "AI is killing creativity, undermining humanity, and completely sabotaging art." I disagree, and I want to explain why.
But first, what is creativity itself?
My definition is this: the conceptualisation and execution of an idea to solve a problem or express oneself. You have an idea for a book and you sit down and write it; you have been creative. You have a problem at work, you brainstorm, mix and match possibilities until something clicks; you have been creative. Creativity is the full arc from idea to execution.
Where ideas come from
To form ideas, you need a curious mind, one exposed to many sources, one that asks questions.
There is truly nothing new under the sun, so every idea must come from somewhere.
For a human mind, that somewhere is the books, films, classes, conversations, and experiences we feed ourselves. A person who reads widely is more likely to have more ideas. A person who reads across diverse subjects has a larger base from which to combine those ideas in unexpected ways. The real challenge, of course, is retrieval. Pulling the right thing from the depths of memory at the right moment is hard, and the AI is far better positioned to do it quickly and accurately.
But here is what we stand to lose if we outsource that struggle entirely: the epiphany. That feeling of a realisation surfacing slowly, something that was forming just out of reach finally breaking through. The burst of satisfaction that comes after a long flirtation with a problem. That is not a flaw in the human process. It is the point of it. The slowness is a feature. It is where meaning lives.
For an AI model, the somewhere is the data it was trained on. When prompted, it retrieves patterns, evaluates probable responses, and generates an answer. It does this faster and more accurately than any human. But notice that word: prompted. I will come back to it.
The execution of ideas
Execution requires two things: skill and tools. You want to compose a song; you need the skill to write music or play an instrument, and the instrument itself. You want to build an original architectural design; you need the skill to construct and the tools to construct with.
Both humans and AI acquire skills from the same general pool of accumulated human knowledge. The AI learns faster. The human takes longer, but builds something in the process, an embodied understanding that shapes how they see problems going forward.
When it comes to tools, the AI has one medium: the computer. Within that medium, it is extraordinarily powerful. It can produce almost anything that can be expressed digitally. The human can learn to use the computer too, and it takes time, but eventually the human figures out how to direct it toward the solutions they want or the expressions they imagine. Just as people once learned Photoshop to make art, or CGI to tell stories in new ways, AI becomes another tool in that lineage.
And this is where I return to the question of whether AI is truly being creative when it synthesises an answer. I said earlier I think not, and here is why.
The human at the top
To generate any output, the AI must be prompted by a human. Someone has to pose the question, frame the problem, decide what matters. At the highest level, there is always a person. The output exists because of a human's original curiosity or intention. In this way, AI is not the creative agent. It is the tool. A very powerful one, but a tool nonetheless.
This does not diminish what it produces. It clarifies who is responsible for it. The person who directs the AI, curates its outputs, and shapes them toward a vision is the author. The question of credit and authorship that AI seems to complicate is actually not that complicated: follow the intention, and you find the human.
A new freedom
What AI does change, and this matters enormously, is who gets to execute their ideas.
For most of history, execution has required either expensive tools, rare technical skills, or both. A person with a vivid imagination but no access to training, equipment, or institutions was largely locked out. Their ideas stayed ideas.
That is changing.
The village kid with a big imagination and a dream to tell African sci-fi stories in their head can now experiment and begin to express that vision with almost nothing but a phone and curiosity.
AI lowers the barrier between having an idea and doing something with it. That is not a threat to creativity. It is the most significant democratisation of creative expression in a generation.
What this means for how we work
We do not need to draw a fixed line on how far we can stretch our tools. We keep generating ideas, developing skills, and choosing which tools to reach for and when. Sometimes the process will be faster, and the human will realise they miss the slower way and return to it. Sometimes they will hand off the parts they care least about so they can linger on the parts they love.
Maybe the filmmaker knows exactly what kind of sound he wants for a scene and generates a rough version quickly with AI, then spends his real energy in the edit. Maybe the sound editor sends the cutting of scenes parts to an AI agent and pours herself into the sonic detail. The overall work is more expressive, not less.
Creativity was never really about suffering through every step alone. It has always been about the idea, the intention, and the will to make something that did not exist before. AI does not touch any of that. It just gives us better tools to get there.