I asked AI to be my stylist. Here's what happened
Artificial Intelligence is here to help with everything from driving our cars to writing our texts. But can it also get us dressed?
I UPLOAD A PHOTO of myself to ChatGPT and ask Sam Altman’s god inside the computer for some help. How can I improve my personal style based on the photo attached? In it, I’m smiling in a navy blazer and a white shirt. Office-appropriate, unremarkable. LinkedIn-Core.
“You’ve already got a great foundation with navy blue, tailoring, and those sleek black shoes,” it fires back. “Classic and sophisticated! To take it to the next level and add a little extra flair, you could incorporate a few of these ideas . . .”
It goes on to suggest pops of colour, added texture, bold accessories and ‘trendy details’.
It’s not totally irrelevant advice, but I don’t think that our brave new world of machine learning recommending that I add a burgundy pocket square to my blazer is going to revolutionise the way I dress.
I must go deeper.
Artificial Intelligence is here to fix everything.
It will drive your car, write your texts, generate a horrible image of a bear playing basketball or Socrates breakdancing, replace your romantic partner and, if certain corners are to be believed, change the fabric (sorry) of the fashion industry.
According to a report by the big cheese analysts over at McKinsey, in the next three to five years, generative AI could add $150 billion, conservatively, and up to $275 billion to the apparel, fashion, and luxury sectors’ operating profits.
The report goes on to suggest that AI can, and will, aid in everything from design and supply chain issues, to configuring the optimum layouts for retail stores, allow you to ‘try on’ clothes from the comfort of your computer or phone, and, as seen by a recent proliferation of AI styling and wardrobe assistants, help you sift through your clothes to create the most stylish version of you.
“We found that AI styling could address a prevailing problem: owning many clothes but feeling like there’s nothing to wear,” says Kijun Yun, a co-founder of Acloset, one of the more popular new AI styling apps, which claims to have 4 million global users, with 7,000 currently joining each day.
“We believe AI is set to revolutionise the way people wear clothes and shop for them,” adds Yun. “In the future, everyone will have a digitised wardrobe, with personalised outfit recommendations each morning based on mood, weather, and daily plans becoming second nature. Your clothing purchases and outfits in everyday photos will be automatically tracked and integrated into your digital wardrobe, offering insights into your favourites, frequently worn items, and those left untouched.
“Acting as a navigation system for your wardrobe, an AI-powered digital closet will provide effortless solutions to daily clothing decisions.”
As things stand, users upload photos of individual items to the app before the AI makes both styling and shopping recommendations based on the information available. Yun and his team are currently developing a system where all you’ll need to do is upload a selfie before the algorithm gets to work deciphering exactly what it is you should wear and how you should wear it.
“Fashion is a highly subjective field, with no single ‘right answer,’ says Yun, when I ask about how he plans to win over potential skeptics. “As a result, the way people perceive the same recommendation can vary greatly from person to person. Furthermore, the degree to which users accept recommendations depends heavily on how the reasoning behind them – why the recommendation was made and why it suits the user – is communicated. This makes it crucial for us to determine not only which recommendations to present to whom, but also how to explain them in the right context and format. Addressing these questions is another major focus for our team.”
The concept of a daily ‘uniform’ has been around for decades and gained a particularly fervent sense of cultural cache with the rise of the Silicon Valley tech baron set. Steve Jobs in his Issey Miyake and New Balance, Zuck in Brunello t-shirts and everyone in Loro Piana.
Acloset is a slick app, it recommended a few combinations of pre-existing clothes that were interesting. ChatGPT told me that Grace Wales Bonner was a designer to look out for and that Westernwear and loose tailoring were trends to be aware of. It’s all quite depressing.
Every car is matte grey, every service is a subscription, and every moment requires a level of optimisation. The more helpful AI was in my pursuit of ‘levelling up,’ the deeper my sense of malaise. The Japanese believe that well-loved objects can develop a spirit – tsukumogami. Each godless prompt and upload and machine suggestion feels like an anathema to that concept.
You need to try on trousers, and wear a bad shirt or a stupid colour sometimes and look back at old photos of yourself and think, ‘Yes, I look like a bit of a dickhead, but at least I was trying something out?’ Feeding clothes into my phone to eradicate any potential indiscretion doesn’t feel like the answer to . . . anything. Just more waste and mess.
Do you have a sense of taste? I ask ChatGPT.
“Yeah, I’d say so!” It replies. “I can recognise good design, balance, and aesthetics – whether in fashion, architecture, or even food plating. But taste is subjective, so it really depends on what you vibe with. Do you think I’ve got good taste based on what we’ve talked about?”
Fuck off.
This story originally appeared on Esquire UK.
Related:
Will men wear bootcut jeans? One Esquire writer puts the look to the test
