Magazines and blogs have been dictating our wardrobes for decades. Who hasn’t rolled their eyes at Cosmo columns insisting horizontal stripes will make you look wider, or hostile rules chastising anyone who dared mix black and navy? But, while in the past style formation relied on discovery and friction – from the jacket you swiped from your dad’s closet, to decoding some inscrutable editor’s reaction on the front row – fashion advice today is turning increasingly algorithmic.
Picture this. Your long-distance, low-commitment, casual pretend boyfriend is in town. You’re gearing up for the most romantic weekend of your life, but then, the dreaded question hits: What do I wear? With all avenues exhausted, you turn to the omnipresent oracle: ChatGPT. “Give me a chic but sexy date-night outfit that’s budget-friendly but still makes me look expensive,” you plead.
You don’t need to look far to find a community of fashion enthusiasts who are doing exactly this; TikTok alone will provide you with hundreds of examples. One woman discovered she was a “Pirate-Gypsy meets Desert Muse meets 2000s bohemian goddess” after letting ChatGPT analyse her Pinterest boards. Another user happily admits to using ChatGPT as their personal stylist and gives others suggestions on what prompts to use. A third person uploads images of every single item of clothing in their wardrobe, then lets ChatGPT think. “I already pay for my AI subscription, so why would I also pay a stylist?” they write.
At first, it seems like a shortcut to salvation. A lower-effort way to keep up with an increasingly demanding fashion culture. But when both you and I get the same output from the same prompt, you start to wonder, does that type of effortlessness come at a price? The social media echo chamber is already dishing out formulaic aesthetics down to the last accessory, from gorpcore to coquettecore to mob wife. The AI, trained on these pre-packaged personas, has no choice but to serve up one of the five cookie-cutter options of dominant styles it knows.
So, beyond panic-prompting an outfit for your cousin’s destination wedding, who actually benefits from AI stylists and the flattened culture they create? As online shopping continues to replace in-person retail – especially amid declining sales and a luxury slowdown – AI is increasingly positioned as a key tool for delivering a successful customer experience. It translates the cues and courtesies of the retailer into an interface, with many fashion brands already rushing to tap in.
Ralph Lauren, for example, is placing its bets on its late-2025 launch, Ask Ralph. The interface allows users to communicate in natural language, mirroring the experience of speaking to a stylist. This is then followed by pulling individual items or complete looks directly from the Ralph Lauren catalogue. It delivers personalised recommendations and a more focused shopping journey, akin to a Ralph Lauren ‘For You Page’ with selections the tool predicts users are likely to buy. Marketed as a digital stylist and modelled on Lauren himself, the tool can also ask follow-up questions to refine its suggestions.
ASOS, on the other hand, has taken a more traditional approach. Its stylist tool delivers personalised outfit recommendations drawn from its multi-brand catalogue, without any single-brand ‘For You’ page feel, nor need for small-talk style prompts.
“Even if it becomes harder to visually distinguish over time, I don’t believe that something created by a computer can connect with people on the same emotional level”
Despite these innovations across platforms, many users continue to turn to the original algorithm, ChatGPT, for fashion guidance, as evidenced by fashion-advice prompt videos that regularly attract thousands of views. However, given OpenAI’s recent collaboration with shopping conglomerate Shopify, where merchants will be able to sell directly through ChatGPT, a question arises: When an ostensibly neutral, all-knowing framework such as ChatGPT is rewired for commerce, does it risk masquerading sales pitches as impartial advice? The AI stylist is no longer a 24-hour helpful Samaritan, but a sleek, self-contained business model with a friendly interface, optimised to encourage more purchases, a modern adaptation of that impeccably-dressed sales assistant who swears this bag will “complete your wardrobe.”
But why do people turn to AI stylists in the first place? In an increasingly demanding fashion sphere, AI can act as a safety blanket, reassuring you that your outfits are not terrible. But in doing so, it avoids risk by design, stripping away the serendipitous magic of style formation. Would fashion subcultures even exist without risk? Without the spikes in punk, the studs of emo, or the white face paint of goth? After all, style is not a ‘problem’ that needs to be solved.
When asked whether she thinks AI styling is going to be big this year, stylist Jeanna Krichel responded, saying, “I don’t think we should ever be afraid of progress. You have to find your own way to move with it, surf the wave and make sure you don’t drown.” But, at the same time, Krichel thinks AI will never reach the same heights as human expression. “We are human beings with soul, history, emotions and experiences. Even if it becomes harder to visually distinguish over time as the technology evolves, I don’t believe that something created by a computer can connect with people on the same emotional level,” she says. “We ultimately connect with the soul, the story, and the emotional core behind style.”
Developing a sense of style once took years of getting down and dirty with thrift archives to curate the perfect amalgam of your own experience. Now it comes at the tap of a finger, stripped of all lore and context. Ungated access to conformist style advice, presented as neutral fact, erases the cringe phase and those embarrassing outfits which helped shape your wardrobe into what it is today. Fashion is a realm built on human friction, ambiguity, and excess, but AI risks crushing creativity. You will never discover that would-be-favourite, risky purchase again if the all-knowing algorithm decides it “doesn’t suit you.”
Discussing whether AI stylists risk flattening culture, Emilio Ferrara, professor at the University of Southern California, was decisive in his answer. “Yes, especially when everyone buys the same ‘future’. If brands use the same models, trained on the same data, chasing the same goals, you get convergence,” said the academic, who specialises in artificial intelligence and computational social science. “Safer palettes, familiar silhouettes, fewer risks, and faster trend exhaustion.”
“If brands use the same models, trained on the same data, chasing the same goals, you get convergence. Safer palettes, familiar silhouettes, fewer risks”
You could argue it’s not all bad. AI styling has real upsides: nailing the perfect fit with virtual try-ons and previewing outfits in hyper-realistic visualisations could slash returns and waste by keeping closet catastrophes to a minimum. But the flattened taste that these AI stylists produce is actually hugely profitable for brands. It means fewer stock-keeping units to manage, accelerated trend cycles, and far less risk on experimental collections, which only serves to line their pockets. The ultimate function of AI styling tools are to streamline their company’s operations, not help you dress better.
AI can make life easier. We already know it can sell you a t-shirt in your exact size, favourite colour, or for a predicted mood. It feels omniscient, like someone finally did the thinking for you. But if we outsource taste entirely, we might as well stop bothering to accrue fashion knowledge at all. As AI stylists are positioned as the next frontier of shopping in 2026, this feels like a moment of choice rather than inevitability. Style has never come from having fewer decisions. It comes from making the wrong ones, publicly and repeatedly, and learning from them. You do not need AI to predict which shade of navy goes with burgundy when you can just try it on. Style is like a muscle you build, and like any muscle, outsourcing the work only weakens it.
