AI Is Changing the Rules of Creativity

There’s a moment, if you spend enough time around working artists, when the subject of AI comes up and the room gets complicated. Not angry, not dismissive — complicated. A graphic designer in Brooklyn might admit she uses Midjourney to rough out client concepts before touching her tablet.

A film composer in Los Angeles will tell you, almost in a whisper, that he ran a melody through an AI arrangement tool last month and the output was genuinely good. Then both of them will go quiet, as if they’ve confessed something they’re not sure how to feel about yet.

That ambivalence is, perhaps, the most honest response to what’s happening right now. AI is not arriving in the creative world with warning sirens. It’s seeping in through side doors — through the tools professionals already use, through the platforms where audiences find new music, through the back rooms of advertising agencies that need twelve logo variations by Thursday morning. And by the time most people notice how much has shifted, a great deal already has.

The numbers are hard to ignore. Research from the Universitat Oberta de Catalunya suggests generative AI could touch roughly a quarter of all tasks across arts, design, entertainment, and media. Spotify, in 2025, quietly pulled over 75 million tracks from its platform — a significant portion of them AI-generated content that had been flooding the ecosystem, diluting royalty pools, and muddying the listening experience for everyone.

The fake band Velvet Sundown racked up more than a million streams before anyone confirmed it was entirely synthetic. On Etsy, human craft sellers started complaining that their listings were disappearing beneath waves of AI-generated product designs. These aren’t dystopian scenarios from a science fiction novel. They’re business news from last year.

It’s still unclear whether the broader public fully understands what’s being disrupted. When Tilly Norwood appeared in September 2025 — a fully AI-generated actress, complete with a promotional reel and an inquiry sent to Hollywood talent agencies — the reaction was swift and visceral. Emily Blunt described the prospect as genuinely frightening. SAG-AFTRA put out a statement pointing out that Norwood had no life experience, no emotion, and was built, at least in part, on the recorded performances of real human actors who never consented to that use.

Musician Stella Hennen noted that Norwood looked uncomfortably like her. Scottish actress Briony Monroe went further, consulting her union over potential personality rights violations. The whole episode had the feeling of a test case — not just for entertainment law, but for something larger and harder to name.

And yet the honest answer isn’t that AI is simply the villain in this story. Throughout the history of art, new technology has consistently sparked the same anxiety before eventually being absorbed, and sometimes celebrated. Oil paint was a technological disruption.

So was the synthesizer. Digital photography made darkroom purists furious and eventually gave us Instagram, which gave millions of people a creative outlet they never had before. Generative AI is doing something similar, though the speed of it is dizzying and the scale is different in ways that matter.

Dominika Weglarz, who spent years researching AI adoption in creative industries for her doctoral thesis at UOC, put it plainly: AI does not replace human creativity, but it can meaningfully enhance it. Her point was practical. When AI tools take over the mechanical, repetitive parts of creative production — the background image searches, the initial layout mockups, the first draft of a product description — they can free up the thinking that actually matters. Critical thinking. Conceptual direction.

The part of the work that requires a person who has lived something. In brainstorming sessions, she noted, AI can generate volume quickly, which paradoxically gives human collaborators more space to evaluate and filter, rather than less. The irony is real: more ideas, faster, sometimes means more room for genuine human judgment, not less.

Still, it would be too easy to stop there. Artist Greg Rutkowski — whose distinctive painterly style took decades to develop — found his name becoming one of the most commonly used prompts in AI image generation.

Anyone could conjure something that looked like his work by typing his name into a text field. He had no say in it, received nothing from it, and suddenly found the market for his distinctive visual voice complicated in ways he hadn’t anticipated. That’s not an abstract IP dispute. That’s a person watching something they built over a lifetime get turned into a publicly available aesthetic.

There’s a sense, watching all of this unfold, that the real argument isn’t about whether AI can produce something that looks like art. It clearly can — sometimes startlingly well. Actress Betty Gilpin, in an open letter responding to Tilly Norwood’s debut, wrote something that feels closer to the actual question: the AI creation looked good, she said, but it looked empty.

It didn’t make her feel like her cells were trading with someone else’s. It made her feel alone. That’s not a technical critique. That’s a description of what art is actually supposed to do — establish contact between one human interior and another.

The legal and economic questions will eventually get sorted, or at least managed. Copyright law will catch up, in its slow and imperfect way. Platforms will develop policies. Unions will negotiate. The harder question — what it means to connect with a piece of work knowing no human made it — doesn’t have an obvious resolution.

It’s possible that audiences will simply adjust, the way they adjusted to digital photography and auto-tuned vocals. It’s also possible that human-made art develops a kind of premium, an authenticity marker, the way handmade furniture commands prices that IKEA never will.

What seems certain is that the creative professionals who will navigate this best are the ones who neither panic nor ignore it. The ones who pick up the tool, figure out what it can and cannot do, and stay honest about the difference. AI is changing the rules of creativity. It is not, at least not yet, writing them alone.