Using AI as a Creative Starting Point (Without Handing Over the Wheel)

There's a lot of noise right now about AI and creativity. Most of it lands in one of two places.

Either AI is going to destroy human creative expression and everyone using it is a sellout. Or AI is a magic wand that does the creative work for you and isn't that convenient.

Both of those miss the point.

I've been using AI tools as part of my creative and business process for a while now. Not to generate finished work and call it mine. Not to replace the part of making that actually matters. But as a genuine thinking partner. A way to move faster, get unstuck, test ideas before I invest too much in them, and challenge my own thinking when I'm too close to something to see it clearly.

AI didn't kill the blank page problem. It helped me solve it.

The hardest part of making something isn't the middle. It's the start.

That moment where you know you want to create something but you're staring at nothing and your brain has decided this is an excellent time to go completely offline. No ideas. No direction. Just the weight of the blank page and the growing suspicion that maybe you don't actually have anything to say.

I've been there more times than I can count. And what I've found is that AI is genuinely useful for exactly this problem.

Not because it gives me the answer. But because it gives me something to react to.

There's a big difference between starting from nothing and starting from something you disagree with. The second one is much easier. Ask AI for three directions you could take a project, and even if all three are wrong, your brain immediately knows why they're wrong and what it actually wants instead. The blank page problem dissolves. You're no longer trying to generate from zero. You're reacting, editing, pushing back.

For a brain that struggles with initiation, that's not a small thing. That's the difference between starting and not starting.

It's a tool. Like any other tool.

A paintbrush doesn't make you a painter. A camera doesn't make you a photographer. Having the fanciest supplies in the room has never been the thing that makes someone an artist.

AI is the same. It's a tool. A genuinely useful one when you know how to use it, and a shortcut to mediocre output when you don't.

The mistake most people make with AI in creative work is treating the output as the destination. You ask for an idea, AI gives you an idea, you take the idea and run with it unchanged. That's not using AI as a creative tool. That's outsourcing the creative process and calling it making.

The raw output from AI is almost never the thing worth keeping. It's the starting material. The thing you react to, push back against, pull one thread from, combine with something else entirely, or use to figure out what you actually wanted in the first place.

That requires judgment. And judgment is the part that's yours.

Learning to use AI well as a creative tool means learning to decide what to keep, what to discard, and what to push further. That's a skill. It takes practice. And the people who develop it end up with a genuinely useful thinking partner. The people who skip that part end up with content that looks like everyone else's.

Where I actually use it

Here's what this looks like in practice for me.

When I have an idea I'm excited about, I'll take it to AI before I invest too much time in it. Not to validate it. To stress-test it. I'll describe the concept and ask for the problems with it, the ways it could fail, the angles I might not have considered. Half the time it confirms I'm on the right track. The other half it surfaces something I'd been quietly avoiding thinking about. Either way, I find out faster than I would have on my own.

When I'm stuck mid-project, I use it to get past the barrier. Not to finish the thing for me. Just to get moving again. Sometimes that looks like asking for three completely different directions I could take something. Sometimes it's asking a question about the work itself. What's this actually about? What's missing? The answer AI gives me is almost never the answer I use. But it unsticks something, and then I can find my own way forward.

When I'm starting something new and the options feel overwhelming, I'll ask for constraints. Give me a color palette. Give me five words to work with. Give me a starting point that's not mine so I have something to push against. For a brain that freezes when faced with too many choices, a starting constraint is genuinely useful.

When I can't name what I'm feeling or what I'm trying to make, I'll describe the mood and ask for language or imagery that matches it. Not to use that language directly. To find the thing I was reaching for but couldn't articulate yet.

And on the business side, I use it as a thinking partner. Testing ideas, working through strategy, challenging assumptions I've gotten too attached to. The same way I'd use a trusted collaborator, except available at 11 pm when the idea won't stop spinning.

The part people get wrong

There's a version of using AI that looks like using AI but is actually just skipping the creative process entirely.

You ask for an idea. You get an idea. You use the idea. Done.

That's not making something. That's ordering from a menu and calling yourself a chef.

The output AI gives you is almost always generic. It's built from patterns across everything that already exists, which means it defaults toward the middle. The expected answer. The most common interpretation. The version of the thing that nobody would argue with and nobody would remember either.

Your job is to take that and make it not that.

Which means developing the judgment to know when something is worth keeping, when it needs to be pushed further, when it's pointing in the right direction but isn't the direction itself, and when it's just wrong and you need to go a different way entirely. That judgment comes from knowing your own work, your own voice, your own standards well enough to recognize when something measures up and when it doesn't.

That's the part AI can't do for you. And it's also the part that makes you an artist rather than someone who has AI.

The people who use AI well in their creative process aren't the ones who are best at prompting. They're the ones who are clearest about what they're actually trying to make. The tool works in proportion to how well you know yourself.

The ethical layer

This wouldn't be an honest conversation without naming it.

There are real issues with AI. The way some tools were trained on creative work without consent. The way AI-generated images have been used to undercut working artists. The way it can be used to flood the internet with content that looks like something but isn't anything. These aren't small concerns, and they don't disappear because the tool is useful.

I'm not going to pretend otherwise.

What I've landed on is this: the ethics of AI in creative work aren't really about the tool itself. They're about how you use it and what you use it for. Using AI to generate images that replace the work of human illustrators is a different thing from using it to brainstorm directions for your own project. Using it to mass-produce generic content is a different thing from using it to get unstuck on a piece that's been sitting unfinished for three weeks.

The line I try to hold is this: AI is in my process. It's not my output.

What I make is still mine. The thinking is mine. The decisions are mine. The voice is mine. AI is part of how I get there, the same way a sketchbook is part of how I get there, or a conversation with someone I trust, or a long walk when something isn't working yet.

You get to decide where your own line is. But I'd encourage you to think about it consciously rather than either avoiding the tool entirely out of guilt or using it without thinking about what you're actually doing.

Start somewhere. That's the whole point.

AI won't make you more creative. It won't fix a creative practice that isn't working or replace the part of making that actually matters. It's not magic, and it's not the enemy.

It's just a tool that's genuinely useful if you know what you're trying to do with it.

If you're curious, start small. Next time you're staring at a blank page, describe what you're trying to make to an AI and ask for three directions you could take it. Don't use any of them directly. Just notice what you want to argue with, what you'd do differently, what it makes you realize you actually wanted in the first place.

That's the whole move. React, push back, decide. Keep what's yours. Discard the rest.

The making is still yours. It always was.

The DIYvinci Community is where I think through a lot of this stuff alongside other creatives who are figuring it out too. If you want a space to talk about your process, share what you're working on, and be around people who understand what it actually takes to make things in a real life, come find us.

It's free. community.diyvinci.com