AI Is Coming for Fashion’s Creative Class

Earlier this year, when Veronica Zhai was preparing to launch her bespoke luxury brand Zhai, she knew her marketing budget would be tight.

Agencies and seasoned brand strategists charged anywhere from $5,000 to $30,000 a month, beyond what she could afford before she’d brought in any real revenue.

So Zhai turned to ChatGPT.

“AI does a lot in terms of the digital marketing effort,” said Zhai, a New-York based designer who previously worked as a trader at JP Morgan. “I’m working with ChatGPT every single day to draft and refine my social posts.”

Bootstrapping designers like Zhai are far from the only ones employing AI this way. Across the fashion industry, brands and retailers are routinely using the technology for everything from building websites to crafting social media captions.

For a time, AI mostly handled routine tasks such as generating product blurbs, sorting inventory or answering customer service chats. But increasingly, it’s stepping into creative roles, too. H&M said in March it would create AI-generated visuals starring digital twins of real models. There are whispers that some brands are exploring the idea of “AI creative directors” — roles built around crafting prompts in tools like Midjourney or Runway to generate visuals at scale.

Such talk usually generates a backlash from fashion creatives: surely AI can’t capture the nuance that comes from a human’s lived experience and cultural knowledge. It’s this spark that separates the work of photographers, stylists, illustrators, PR strategists, writers and brand architects from AI’s output.

Will that always be true though? As AI gets better and better, it’s undeniably taking on more tasks once seen as inextricably human. But creatives still have a powerful edge — if they adapt.

That can mean running toward AI, not away from it, and treating it as one tool among many. The key to staying competitive lies in doubling down on the “distinctly human skills” machines can’t replicate like emotional intelligence, storytelling, cultural fluency, authenticity, craftsmanship and the ability to build connections, said Yvonne Pengue, founder and director of Spot On Minds, a London-based executive search firm and consultancy focused on luxury brands.

“I understand the worry, and we are always worried about the unknown,” Pengue said. “But to stay competitive, people need to push themselves to really want to bring that added value.”

Defining the Human Spark

While the hopeful refrain — that AI may be fast, but it can’t replicate the “human spark” — is increasingly being put to the test, experts argue that AI remains unlikely to fully replace human labour anytime soon. Instead, it is reshaping how companies value it, forcing both employers and creative talent to define what that “spark” actually means.

The answer is emerging through AI itself. As creatives and brands push the limits of these tools, they’re also exposing their boundaries — and, in doing so, helping to clarify what remains uniquely human: the nuanced but distinctive blend of emotional intelligence, intuition, personal connection, and cultural experience that gives AI something to build on.

For South African fashion and arts writer and photographer Nabeela Karim, those boundaries came into focus when she began working directly with AI. After being laid off in September 2024 from her role as PR and communications officer at a luxury-fashion focused marketing agency, she re-entered the job market to find PR copywriting roles replaced with listings for “AI editors” — jobs focused on cleaning up machine-generated copy. She reluctantly accepted one, only to find the results frustrating for both her and the client, a nonprofit focused on humanitarian issues.

“They kept telling me they didn’t like the copy,” she said. “It still feels generic. It doesn’t feel like it’s written by people.”

She soon realised the real issue wasn’t grammar or polish — but rather voice and perspective, she said. Instead of relying on AI to draft the content, Karim began writing it herself, drawing on her knowledge of South African culture, photojournalism, and storytelling to create more authentic material — then used AI only to clean up typos and tighten the language. In doing so, she helped the organisation recognise it wasn’t just an editor it needed, but a human voice.

Later, she brought that insight to a fashion designer struggling with website copy and convinced her to scrap plans to use AI in favour of hiring a human writer — Karim herself. “It worked,” she said. She landed a paying gig creating the site’s content.

Zhai, meanwhile, has been using AI as an assistant for tasks like content ideation, social media posts, and even building her brand’s website. But it hasn’t replaced her need for human collaborators. She recently hired two interns — not for their technical skills, but for something less tangible.

“I care less about their [hard] skills because AI can do a lot of execution,” Zhai said of the interns, who help with copywriting and more hands-on tasks like preparing for trunk shows. “I care more about their intuition, their attitude, their humanity and connection with me, their concept and strategy.”

As her brand grows, she said she plans to “pay up” for one “very elite senior brand creative” who has a few other things AI doesn’t: “the network and the knowledge … and has actually [built a brand] before,” she said.

Working With AI

The bottom line: when brands hire for creative roles, they’re looking for people who bring original ideas to the table. But in a head-to-head scenario, the creative who uses AI to enhance their process and/or end results will have the edge, said Cyril Foiret, founder of Maison Meta, a generative AI agency that builds custom tools for brands like Moncler and Dolce & Gabbana.

As AI becomes more embedded in creative workflows, best practices are also starting to emerge — especially around where human creativity ends and machine input begins. Brands encouraging artists, photographers and designers to use AI are also beginning to put up guardrails: with some now requiring talent to “show their work” by logging the steps they took (often the prompts they used) to demonstrate the creative process, Foiret said.

“In Europe when you do AI work, it has to be announced [and] you have to be able to prove the amount of human work,” he said.

Even in places like the US, where regulation is for the most part nonexistent, Foiret believes low-effort creativity that stems from simply plugging in “a few artist names as inspiration” and pressing a button — without deeper thought or skill — will eventually weed itself out.

Things like craftsmanship (sewing and handmaking items like luxury handbags or garments) or photography, where a sharp eye, emotional intelligence, and timing matter, aren’t being replaced by machines. But AI can supercharge those processes, helping creatives ideate, drill down on ideas, and sharpen their results, experts say.

“As an artist or a designer, before, you used to see things in museums or online … and then design from that,” he said. ” “AI now has those [ideas] inside of it already.”

Designers looking to protect their vision while using AI to enhance it are also exploring ways to maintain creative control — like working off the cloud (i.e., on local servers rather than shared networks), which can help safeguard intellectual property.

Norma Kamali, for example, worked with Maison Meta to train a custom AI model on her design language, with the goal of preserving the artistry that defined her career even after she steps back.

“Norma is the epitome of jumping in — testing, learning by herself,” Foiret said.

As more brands adopt AI, creative professionals will be expected to bring not just raw talent, but also the ability to work with these tools. Prompt engineering — the skill of guiding AI to generate useful results — will be increasingly valuable, Pengue said.

“AI can produce more ideas, but there has to be a human element to evolving it,” she said. “The [creatives] need to become masters in using the prompts correctly to generate the right outcome, and to constantly nurture AI and guide it so that it can help us.”