Creativity is Scarce Now. Hear Me Out?

I have been using AI for a while now, and honestly, I think it is one of the most interesting technologies we have ever built. As a developer, I use it to explore ideas, debug code, understand things faster, and get unstuck when I have been staring at the same problem for too long. I also spend a lot of time designing, so I have seen how quickly AI can turn an idea into something visual. It is genuinely useful, and I am not against it at all.

But lately, I have been thinking about something that goes deeper than jobs or productivity. I think we are slowly losing the habit of thinking and reasoning creatively, and the real danger is not that AI is becoming more creative but that we are letting AI become the source of creative direction rather than just an implementation tool.

Today, if you need an idea, you can ask AI for one. If you need a design, you can generate one. If you need a website, you can get both the design and code generated for you. If something you wrote does not sound right, AI can rewrite it. If you are stuck on a project, you can ask it what to do next. None of this is necessarily bad, but the problem starts when we outsource the thinking itself. When we stop asking what should I create? and start asking what should AI create for me? we trade our creative muscles for convenience. The deeper issue is not that AI makes us dependent, though that can happen, nor that AI-generated work tends to look the same, though it often does. The real risk is that we bypass the very process that develops our creative judgment.

There used to be a certain amount of struggle involved in making things. You would have an idea, build the first version, realise it was terrible, change it, break something else, fix that, and slowly get better. Design works the same way: you try different layouts, colours, typography and interactions until something finally feels right. That process can be frustrating, but it is also how you develop an eye for design and learn how to solve problems. Now, it is very easy to skip most of that process and go straight to something that already looks finished. When you outsource the thinking to AI, you never get the feedback loop that sharpens your instincts.

That is the part that worries me. A developer can generate an implementation without fully understanding the problem, a designer can generate twenty interfaces without really developing a sense of composition, and a student can get an answer without spending enough time understanding the question. The final result might look impressive, but that does not necessarily mean the person who made it has become better at making things. Worse, we might start presenting work that looks finished without actually being ours.

I saw this clearly in a recent conversation with a developer friend. He is a full‑stack developer with five years of experience, has trained other developers, and is building his own credit scoring infrastructure. But when I looked at his portfolio, it did not show any of that. It looked generic, like a guy that just started learning CSS after HTML, and it screamed AI. The portfolio failed to communicate his actual ability, not because he wasn't capable, but because he had let AI dictate the design direction rather than using it to implement his own vision.

He was offended at first and thought the critic just did not know how to talk, but I told him it was not about that. Investors and recruiters will be even harsher; they will not care about how many years you claim to have but only about what they can see. His portfolio was not for him to admire but for the people hiring him, and if it looks like every other template on the internet, they will assume he is like every other developer. The issue wasn't that his portfolio was "bad" in an absolute sense; it was that it didn't tell his story.

I suggested he take time to rebuild it from scratch: sketch with pen and paper, browse Pinterest and Cosmos for inspiration, and think carefully about every detail. I told him that AI should only implement his creativity, not create it, and that the bulk of the work should be his own thinking, planning, and designing, while the AI handles the repetitive, mechanical parts. The exact division isn't important; what matters is that he, not the machine, owns the creative direction.

He pushed back at first, saying he liked simple and classic, but we talked about the difference between intentional simplicity and the kind that comes from not having put much thought into it. Simplicity is powerful when it's deliberate; it's hollow when it's just the default output of a model trained on everyone else's learned patterns. I showed him another portfolio to compare, and he eventually started redesigning. He split things into pages, added more moderation, and began to enjoy the process. But I could still tell AI was involved because the shadows, the layout, and the overall feel were too predictable. It was fine, but it wasn't uniquely his; the AI was still dictating the creative direction, not just doing the heavy lifting.

That is the core issue. When AI becomes the source of creative direction, every portfolio starts to look the same, every design follows the same smooth gradients, box shadows, and predictable spacing. It is not that those things are bad in themselves, but they come from learned patterns drawn from thousands of existing designs rather than from your own reasoning about what your work should communicate. Your unique perspective disappears into the noise. And more importantly, you never develop the muscle to make those decisions yourself, so the next time you face a blank page, you reach for the AI prompt again instead of reaching for your own ideas.

The conversation was frustrating for him, but it was also a gift because he realised that his portfolio did not prove his five years of experience. That is the real cost of letting AI drive creativity: it is not that AI is bad, but when you let it create for you instead of with you, you lose the struggle that makes you better and you lose the fingerprints that make your work unmistakably yours.

I do not want less AI; I want better humans using AI. I want it to help us build things we could not have built alone, not replace the part of us that comes up with the idea in the first place. Code will keep getting easier to generate, designs will become easier to generate, writing will become easier to generate, and eventually even generic ideas will become cheap. So maybe the valuable thing will not be who can generate the most but the person who can look at a problem and see something everyone else missed, the person whose portfolio does not just show work but shows them.

Creativity is not disappearing because AI is becoming creative; it becomes scarce when we stop practicing it ourselves. And as someone who loves both engineering and design, I think that would be a much bigger loss than AI taking a few jobs. The real loss is us forgetting how to think for ourselves.