There’s a moment every creative knows. You sit down to work and everything goes quiet in the worst way. The page is blank. The canvas is empty. The DAW is open and the cursor is blinking at you like it’s personally offended you showed up. You have the idea, somewhere, you know it’s in there, but the distance between what you feel and what you can actually make feels enormous. So you close the laptop. You tell yourself you’ll come back to it. You don’t always come back to it.
That moment hasn’t gone away. But something has changed.
There’s a conversation happening right now in creative circles, and it’s honestly overdue. Not the tired one about whether AI will replace artists, writers, and musicians, because that conversation was always more fear than fact. The real conversation is about what happens when a creative who actually knows their craft sits down with an AI that’s genuinely gotten good at asking the right questions. What happens when the blank page starts talking back? Not to write for you. Not to think for you. But to push you further into your own ideas than you could get alone.
That’s where we are right now. And if you’re a creative who’s been watching this from the sidelines, thinking it’s not for you, thinking it’s for tech people or people who’ve already given up on being real artists, this is for you specifically.
Claude and Gemini have crossed a threshold that not enough people are talking about. The outputs from both have started to feel different, less like autocomplete on steroids and more like something that understands tone, subtext, and the weight of what you’re trying to say. They’re not perfect. But they’re asking better questions, catching inconsistencies you’d miss in a third read, and responding to creative direction in a way that actually moves the work forward. That’s new. And if you use it right, it’s genuinely useful.
The keyword there is right.
Here’s the mistake almost every creative makes the first time they sit down with AI. They treat it like a vending machine. You put in a request, you expect output, and when the output is generic and flat and sounds like nothing you’d ever actually make, you decide AI isn’t for creatives after all. You walk away. You tell your friends it doesn’t work.
But here’s the thing. You walked up to a stranger, handed them a blank piece of paper, said “write me a story,” and then got annoyed when they handed you something that had nothing to do with you. That’s not an AI problem. That’s a context problem.
Think about how you learn to cook a new dish. Not the dish you’ve made a hundred times, but something new, something that requires actual technique. You don’t just walk into the kitchen and wing it. You read the recipe. You understand what’s happening at each step and why, not just what to do, but what you’re trying to accomplish. You taste as you go. You adjust the seasoning. The dish doesn’t cook itself and it doesn’t cook because you yelled at the pan. You bring the flavor. The kitchen just gives you the tools to develop it.
AI is the kitchen. You are still the chef.
Context is everything, and it’s not a small thing. It’s not a tip or a trick, it’s the entire foundation of what makes working with AI actually work. When you give AI a writing sample, you’re not just giving it words, you’re giving it your rhythm, your voice, your instinct for where a sentence should breathe and where it should punch. When you give it a mood reference for a track, you’re telling it what emotion lives in the space between the notes. When you give it a character backstory, you’re not just filling in a form, you’re building a person, and AI can only understand that person as deeply as you’re willing to explain them.
Have you ever gotten a response from AI that felt completely off? Cold, generic, nothing like what you had in your head? Odds are you gave it nothing to work with. Not because you were lazy, but because nobody told you that the input is the work. Not the output. The input.
So before you ever ask AI to help you create anything, you have to build the foundation. This is the development stage, and it’s not extra work, it’s the work you’d be doing anyway, just now you’re doing it with a collaborator instead of alone in your head.
For writers, that means character backstories before a single scene is drafted. It means knowing your plot arc, your genre, your tone, what books live in the same neighborhood as yours. It means being able to answer, “What does your protagonist want that they’d never admit out loud?” before you ask AI to write a single line of dialogue. Because if you can’t answer that, the character isn’t ready, and no AI in the world can build a real person out of a name and a vague description.
For musicians, the development stage is the hook idea, the rhythm feel, the emotional story behind why this song needs to exist. It’s the three artists who live somewhere in your sound and why. It’s the tempo energy, whether this track opens slowly and builds or hits immediately. It’s the lyrical theme, what is this song actually about under the surface, not the subject, the feeling. That’s the difference.
For visual artists, it’s the style references and the color mood. It’s being able to say what you want someone to feel when they stand in front of this piece, not just what they see, but what happens inside them. That’s context. That’s what AI needs to actually help you.
None of this is new work. This is the thinking you’d do anyway if you were starting a serious project. You’re just externalizing it now, making it visible, giving it to a collaborator who can respond to it. The difference is that instead of all of that living only in your head, it’s in the conversation. And that means AI can reference it, push back on it, and help you find the gaps you didn’t know were there.
That last part is where things get genuinely interesting.
We’ve been building a book together. And what made that process click wasn’t the moments when AI generated something impressive. It was the moments when AI started asking questions. Good ones. The kind of questions that made me think harder about choices I’d already made.
It asked about the characters in ways that went deeper than surface description. What does this person believe about loyalty? What happened to them before chapter one that they haven’t told anyone? Where does their confidence actually come from, and is any of it earned? Those questions weren’t just useful, they were clarifying. They forced decisions that made the characters more real, more specific, more interesting to follow.
Then it started asking about the story architecture. Where does the reader need to feel safe before you pull the rug? Where should the twist land to hurt the most? Is this plot turn earned by what came before it, or does the reader feel cheated? AI wasn’t making those decisions. But it was asking the questions that forced me to make better ones.
Imagine this. You describe a character to AI. You explain who they are, what they want, where they’re from. And AI asks you, “What does this character want that they’d never admit out loud, even to themselves?” You pause. You think about it. And suddenly you know something about your character that you didn’t know five minutes ago, something that changes how every scene with them will feel. That’s not AI creating your character. That’s AI helping you discover who your character already was.
That’s the magic of the first draft as a conversation rather than a command.
Because the first draft isn’t the product. The demo isn’t the album. The pencil sketch isn’t the painting. They’re tests. They’re proof of concept. They’re the moment where you find out whether the foundation holds or whether there are cracks you need to go back and fix. And when you approach them that way, with AI as the questioning collaborator rather than the content generator, you end up with something much closer to what you actually wanted.
Prompt structure matters here more than most people realize. “Write me a chapter” is a request. “Write this chapter in the voice of someone who’s trying to sound calm but is terrified, keep the sentences short when the tension is high and let them sprawl when the character thinks they’re safe” is a direction. There’s a real difference between the two. One asks AI to invent. The other asks AI to execute something you’ve already defined. The second one will always get you closer.
It’s a little weird, honestly, to have AI push back on your creative choices. To have it say, in so many words, this plot turn doesn’t feel earned, or this character’s motivation isn’t consistent with what you told me about them in the beginning. Your first instinct might be to argue. Maybe you should, sometimes your instinct is right and AI is missing something. But sometimes your first instinct isn’t your best instinct. And having something push back on you, something that has no ego in the outcome, no attachment to the choices you’ve already made, can actually move the work in directions you wouldn’t have gotten to alone.
That’s where iteration lives. And iteration is where good becomes great.
The gap between “close to what I wanted” and “exactly what I wanted” is almost always closed through rounds of conversation, not a single perfect prompt. Think of it like layers rather than rewrites. You’re not throwing out what you built, you’re going deeper into it. The first round of prompts builds the architecture. The second round tests it. The third round refines the details. Each conversation builds on the last, adding specificity, adding texture, adding the kind of nuance that makes something feel real rather than manufactured.
With the book, the plot changes didn’t happen in one session. The thrills and turns and the conclusion, those came out of rounds of questions and answers that got more specific each time. One prompt for tone. Another for structure. A separate one for a specific scene that needed to hit differently. You don’t ask AI to do everything at once any more than you’d try to cook every component of a complex dish in the same pan at the same temperature. You work the process. Each piece gets its own attention.
And through all of it, you’re still the artist. That never changes.
The output will only ever be as interesting as what you put in. Bad inputs get generic outputs, that’s not a bug, it’s physics. But a creative who knows their craft, who knows their story, who has something real to say and can articulate it, even roughly, even imperfectly, that person gets something back from AI that actually moves them.
Think about two musicians sitting down with the same tool. One types “make me a hip hop track.” The other sends a voice memo of a melody idea they hummed into their phone at 2am, writes three paragraphs about the emotional story behind the hook, names three artists whose energy lives somewhere in the DNA of this song, and describes the lyrical direction, what this song is really about underneath the surface. Same tool. Completely different results. The question isn’t what AI can do. The question is what you bring to it. What’s different between those two musicians? Everything. And nothing about the tool.
Creativity has always been about having something to say. AI just helps you figure out how to say it faster, and sometimes, more clearly than you could alone.
Now think about this. Every generation of creative giants had access to something others didn’t fully understand yet. The printing press. The electric guitar. The recording studio. The synthesizer. The internet. Each time, there were people who used the new thing to say something the world hadn’t heard before, and there were people who dismissed it as not real art, not real music, not the right way to create.
Shakespeare had the Globe Theatre and the English language at its most elastic and alive. The Beatles had Abbey Road and producers willing to try anything, including putting orchestras in rock songs and recording albums backwards. Stephen King had a typewriter, a laundromat job, a wife who pulled his manuscript out of a trash can when he thought it was worthless, and stories that would not leave him alone. J.K. Rowling had a napkin and a train delay. Eminem had Detroit and a hunger that cannot be coached or manufactured. Ed Sheeran had a loop pedal, a bedroom, and the willingness to play anywhere anyone would let him in.
None of them waited for permission. None of them had the perfect setup. None of them created in ideal conditions. They had something to say and they used whatever was in front of them to say it.
Now think about who’s sitting somewhere right now. Someone with a character living in their head so vividly they can tell you what that character eats for breakfast, how they walk into a room, what they’re afraid of. Someone who’s been carrying that story for years but the blank page always felt too big, the process too complicated, the gatekeepers too many. Or a musician who can hum a melody that makes people stop talking, but has never been able to afford a studio or learn to read music formally. Or a visual artist with an idea that doesn’t fit any genre they’ve been taught, something genuinely new, but they’ve never had the technical training to execute it the way they see it.
Those people exist right now. They’re real. And the thing that was standing between them and the work was never talent. It was access. Access to the right tools, the right questions, the right process.
What if the only thing standing between the world and the next great story was knowing how to have the right conversation?
That’s not a small thing. That’s everything.
The next Shakespeare might be someone who grew up speaking three languages and writes in the spaces between them. The next Stephen King might be a night shift worker in Phoenix who’s been typing chapters on their phone during breaks for two years. The next Beatles might be a bedroom producer in São Paulo who can hear five tracks in their head simultaneously but never had a band. The next J.K. Rowling might be someone who failed every English class they ever took but never stopped telling stories to anyone who would listen.
They’re not waiting for the right moment or the right setup. They just needed the kitchen.
The creatives who learn to work with AI now, who understand that context is the whole game, who approach the process with intention and craft and something real to say, they’re not getting ahead of a trend. They might be writing the next book that someone reads at 2am and cannot put down. They might be making the record that defines how a generation understood themselves. They might be painting something that makes a stranger in a museum feel, for a moment, completely seen.
The tool is ready. The question is whether you are.
Start before you feel ready. Build the foundation first, spend time on your characters, your hook, your vision, whatever the work requires. Give the AI everything, your samples, your references, your story, your intention. Ask it to ask you questions before you ask it to produce anything. Let the first draft be what it is, a test, not a judgment. Iterate. Go deeper. Keep going until what you’re making starts to feel like yours.
Because it is yours. It always was.
If any of this lands for you, if you’ve got a creative project sitting in a drawer or a notes app or the back of your mind that just needed a new way in, I want to hear about it. Share your experience, tell me what you’re working on, tag me at @iamcezarmoreno and let’s talk about what you’re building. There’s more where this came from at cezarmoreno.com, where I write about technology, creativity, and the intersection of the two for people who want to understand what’s actually happening, not just what the headlines say.
