Do robots dream of Prada? How artificial intelligence is reprogramming fashion | Artificial intelligence
Who do you turn to when you can’t decide what to wear? Your best friend, maybe. Instagram, probably. People like me and magazines like this one (hopefully). But soon, perhaps, it will be none of the above. Instead, you will try on an outfit, turn to a wall-mounted, five megapixel camera with front lighting and dual-antennae wifi connectivity, ask, “Alexa, how do I look?” and within a few seconds the 1.6 watt speaker will deliver the data-driven, empirically-founded assessment.
The Echo Look is Amazon’s first “style assistant”, recently rolled out across the US after an invite-only soft launch. No UK launch date is set, but the technology – which analyses your outfit through a combination of algorithms and (human) “fashion specialists” – is set to revolutionise what technology means to style. Just four years ago, the cutting edge of technology in fashion was Tommy Hilfiger’s solar-powered phone-charging jacket. Horse-and-cart stuff, compared with what is going to happen to fashion next.
The real point of fashion isn’t the fabric or the clothes themselves; it is how we think and feel about those clothes. And it is this human, emotional part of fashion – style, if you like – on which artificial intelligence now sets its sights. Stitch Fix is a online personal styling service which sends its 2.7 million active American clients suggestion boxes of clothes chosen by cross-referencing a client’s stated preferences with the recent purchases of others of similar age and demographic. Matchesfashion.com is experimenting with personalised 3D avatars that will be able to “try on” digital samples so that you can see how the shape and size will work on your body. Net-a-Porter is trialling technology that scans your data for information on forthcoming trips and events, and tailors its suggestions accordingly.
But can there really be an algorithm for style? Surely not. In 2003, Kate Moss found a lemon-yellow 50s chiffon dress in Lily et Cie, a vintage store in Beverly Hills, and wore it to a dinner at New York fashion week, where the entire room fell in love with it and a million copycat versions were born. The dress wasn’t in keeping with that season’s catwalk trends, or colours, but it was somehow absolutely right for that moment. I was there, the dress remains seared on my retina to this day, and it felt like serendipity, like magic. How would an algorithm replace the je ne sais quoi of Kate Moss? Or Jane Birkin in the south of France with Serge, or Bianca Jagger at Studio 54? This is a quantum leap from the frying pan you bought last month that’s still following you around the internet.
The metric of a style algorithm based on likes, whether fed to you as feedback on your selfies or as a subscription box of suggested autumn clothes, will steer you towards a polished, palatable, mainstream look. “If the algorithm is based on mass approval, it is not going to propose you wear a weird top with one sleeve,” says Alistair O’Neill, professor of fashion at Central St Martins. “It’s going to knock the edges off your preferences and guide you towards an aesthetic that is sort of ambient.” Early users of the Echo Look reported that it scored navy and muted colours higher than brighter shades, and gave the thumbs up to Instagram-approved styling accents such as popped collars and rolled-up shirtsleeves.
In the 21st century, technology is defining our taste. Ten years after its launch, Airbnb is not just a platform to rent somewhere to stay, but a silent tastemaker which has drawn a template for how a desirable home should look. White or bright accent walls, raw wood, Nespresso machines, Eames chairs, patterned rugs on bare floors, open shelving, Scandi-chic, the industrial look, and a minimal version of mid-century were characteristics that Kyle Chayka identified as the Airbnb “look” two years ago. Standardisation evolved organically, as would-be hosts copied the look of the most popular spaces on the site. And then renters taken by the bare Edison lightbulbs and gallery walls of black and white photography in homes they stayed in while on holiday began to bring the look into their own homes. It is easy to imagine a similar process taking place in our wardrobes, once facsimile style advice is being beamed into each of our homes.
That would be less an algorithm for style than an algorithm for killing off style, though. “You only have to look at the interest in the V&A’s show about Frida Kahlo this year to see that people really value style as a form of self-creation,” O’Neill says. The subversive, iconoclastic, individual aspect of fashion is important not just to its cultural weight, but to its commercial clout. The power of fashion to make us spend is strongest not when we are presented with another version of the type of pencil skirt we already have and like, but the moment when we see, say, a kilt, and realise that, despite never having wanted one before, we simply have to have one right away.
“Fashion needs audacity,” says Simon Doonan, fashion writer and consultant. “Look at what has happened at Gucci, which Alessandro Michele has reinvented. When he took over, Gucci was quite conservative. If he had tested his crazy ideas against the data about what Gucci clients were buying, there would have been smoke coming out of the computer. And yet somehow it worked. It was his gut instinct and for whatever reason the powers that be were brave enough to go along with it. And here I am now, standing here talking to you wearing silk Gucci slippers with cats embroidered on them.”
But perhaps the style algorithms of the future could be programmed to surprise. Brad Klingenberg, vice-president of Stitch Fix, states that the aim is to “delight” clients, rather than just please them, suggesting an element of the unexpected. (Stitch Fix, like the Echo Look, is currently guided by people as well as data. “We rely on our human stylists to empathise with clients. For example, when a client writes in to her stylists that she needs something to wear to her ex-boyfriend’s wedding, only a human can understand the gravitas of that request,” Klingenberg says.)
To futurists such as Sophie Hackford, we are wrong to romanticise the way fashion works now. “Online shopping as we know it is a lousy experience, because you are essentially looking at an inventory. One day in the future you will be sitting on your sofa next to a virtual Diana Vreeland, or Alexa Chung, who will be talking you through the selection of virtual clothes you can see being modelled in front of you, and it will seem so funny that we once scrolled through two-dimensional skirts on the internet,” she says. Retailers are already experimenting with incorporating data from your calendar – about a future trip, and what the weather forecast is for that location, for instance – into what gets served up as cookies. Artificial intelligence could sprinkle fairy dust on the online shopping experience, so that instead of scrolling through a hundred skirts, you are matched with one you fall in love with. Sandrine Deveaux of ecommerce unicorn Farfetch is working on a “Store of the Future” which hopes to seize the momentum with a new, better customer experience. “The entire industry is focused right now on what the consumer-facing aspect of AI in retail will be. It has to be something meaningful.”
The robots are not necessarily the bad guys. Artificial intelligence could hold the key to making fashion more sustainable. “We are producing too much clothing and throwing away too much clothing,” says Matthew Drinkwater, head of the Fashion Innovation Agency at the London College of Fashion. “The retail model needs to change. AI makes it possible to adjust manufacture in real time, responding to customer design as it happens, so that waste is minimised.” The opportunities for personalisation – from monograms to bespoke tailoring using 3D measurements taken online – hold the promise of clothes that we will value more, and wear for longer.
And there’s more. An artificial intelligence takeover of the power traditionally held by the population of magazine mastheads and the fashion week front row to anoint the “best-dressed” could bring about a democratic revolution in an elitist industry. That the industry is still riven with snobbery and unconscious bias – or worse – about skin colour and body shape is evidenced by the way so-called “streetstyle” galleries on fashion websites, and the upper echelons of the “influencer” world, are dominated by thin white women. Algorithms could be used to avoid the bias and snobberies that hold fashion back. The streetstyle photos selected for magazine websites, for instance, tend to follow a very specific seasonally-ordained look. Machines programmed by humans are known to have their own biases, but an algorithm – programmed into CCTV in cities around the world – might one day do a better job of finding interesting-looking people.
At the heart of our unease about artificial intelligence – not unique to fashion – is a disquiet about the changing power dynamic between human intelligence and the artificial kind. We sense the robots creeping up on us, we imagine them breathing down our necks (unlikely, what with them having no need of respiration, but still) and we worry about how we will compete. And the more artificial intelligence advances into those areas of our thinking that we experience as creative and emotional, the more spooked we get. Artificial intelligence already guides your car the fastest route home; it probably won’t be long until that mapping app communicates with your home hub to put the kettle, music and lights on for your arrival, just like your partner or flatmate might do. That is intelligence, but it will feel a lot like affection, which we think of as a human-to-human interaction. In the same way, algorithms that know your spending power and established habits already manipulate what you will see if you search online for, say, white trainers. But if one day soon you get dressed in the morning and your phone beeps to tell you that your look is lame, what will that feel like? Cyberbullying?
Not so long as we programme the robots to be kind, Doonan says. “I personally don’t like the idea that there’s a right or wrong way to dress. I have a Moschino jacket that says on the back ‘Good Taste Does Not Exist’ and I believe that. But the reality is that many people are very insecure about how they look and they want stuff to wear that helps them feel confident. When I talk to customers at Barneys about how they decide what to wear, a phrase that comes up a lot – particularly among men – is the need to ‘get it right’. So I am sympathetic to some kind of mechanism that reduces anxiety.”
Perhaps, anyway, we are more like the robots than we like think. “The majority of people have already developed an algorithm for style, even if they don’t think of it like that,” says Simon Lock, founder and CEO of Ordre, which offers fashion buyers a digital, streamlined alternative to physical showrooms. “For instance, I wear black and white, a slim fit silhouette, always Thom Browne brogues. Essentially, the eye captures a look and the brain informs the wearer whether you like it or not based on history and personal taste. Artificial intelligence is perfectly suited to perform this role for us.”
Today’s teenagers have an ever more porous boundary between their IRL and online selves, with relationships – even romantic ones – sometimes conducted entirely via phones. (FaceTime, but not face time.) They have developed what Bia Bezamat, innovation editor at online news site TheCurrent, calls a “hybrid identity”. And it’s not just gamers and kids who have avatars. Isn’t the Instagram or Facebook version of yourself – the one who has more covetable holidays, better behaved children and sassier one-liners than actual you – a kind of avatar, too?
The boundaries will blur even more, the futurists say, once the online version of you is able to operate independently, a kind of digital alter ego to whom you can delegate. “I think it’s inevitable that pretty soon we will each be represented in the digital sphere by an avatar,” Hackford says. “Your avatar will check when your parking permit needs updating. It will compare available prices on everything you want to buy. It will sit in the hold queue on the phone to buy a train ticket. It will do all the things that technology does better than you can and allow you more time for being human.”
But right now, it seems we want to be more robot instead. One of autumn’s key catwalk trends, as seen at Alexander Wang, looks a lot like The Matrix, the 1999 sci-fi blockbuster set in a future in which simulated reality has cannibalised human experience. Costume designer Kym Barrett devised the oil-shine black coats to characterise what she called “a shadow world where people would disappear and reappear… as if you were tricked. Is she there? Is she not there?” This dystopian vision has been enthusiastically adopted by Bella and Gigi Hadid and Kendall and Kylie Jenner, four of the most powerful young women in the social media landscape.
Long before this season’s Matrix trend, the widespread use of Photoshop had begun to shift our image of cover-girl perfection from that of a beautiful, real human being to that of a digitised version of human beauty, with impossibly even skin tone and unnaturally symmetrical features. In the last two years, the prettifying face filters that began on Snapchat have spread to Instagram Stories. These enlarge your eyes and lips, plump your cheekbones into a heart shape, replace black pupils with a flash of silver. The effect is deliberately unreal: more cartoon character than supermodel. Imitation, they say, is the sincerest form of flattery. Some things never change, and we all want to look like the cool kids. It’s just that this time around, the new kids on the block are robots.