A few days ago, I attended a lunch for writers and creatives. I was seated on a table with mostly strangers: a film director, a podcaster, a technology and finance journalist, plus one beauty editor (me). The lunch was loosely based on women establishing careers in creative fields against the odds. After the second course (and a second glass of wine), our conversation finally made its way to the topic of Artificial Intelligence. It started optimistically: “A.I. breeds innovation, encourages new ideals and ideas…” But after a beat, the tech journalist admitted she felt humanity was actually “descending straight into a monoculture hellscape”. We laughed, but it was tarnished with nervousness.
It’s true that we’re living in a time where things feel strange. Our thoughts, ideas, quirks, standards and passions are under threat. ChatGPT launched towards the end of 2022. It now has an estimated one billion active monthly users. The global population is around eight billion, with around 75 per cent of us connected to the internet in some way. A lot of people are talking to Chat (not to mention Claude, Perplexity, Gemini, etc). Like, a lot.
A.I. is impacting, or will impact, every corner of our lives. In good ways, and in bad ways. Beauty, as an industry, a zeitgeist and a reflection of culture, is no different. Since that lunch, I’ve been ruminating and decided there’s three main areas of the beauty industry that will be impacted by this rapid technological proliferation: what we consume within beauty, how we consume it, and finally how things get made. Broadly, this bleeds into everything: formulation science, distribution, customer service, creative campaigns and marketing collateral, influence, information aggregation, what we consider expertise (and more importantly, what we consider to be true). Also, what we deem to be beautiful. The latter is perhaps the most perilous.
All of this has created a crisis of meaning and significance for brands, and a crisis of equivocation for consumers. Image production is limitless; information is churned and served at lightning speed with no critical analysis. No one is accountable, either. Different interpretations of what makes something or someone beautiful are becoming more and more homogenised. Pessimists (realists?) would say that an A.I.-affected beauty industry is on track to be a dumpster fire of nothingness and everything-ness simultaneously. It’s difficult to say.
Back in March of this year, The Business of Fashion (BOF) released data pertaining to the most recommended beauty brands by A.I. Within skincare, Paula’s Choice was number one, followed by CeraVe, La Roche-Posay, The Ordinary, Drunk Elephant, Kiehl’s, Olay, Tatcha, Skinceuticals and then Neutrogena in that order. Makeup was as follows: NARS, Maybelline New York, Urban Decay, Rare Beauty, L’Oréal, e.l.f, Too Faced, Tarte, NYX and Anastasia Beverly Hills. Every brand mentioned is owned by a conglomerate.
“We are simultaneously living in a time where acne, stray hairs, crooked teeth and folded wrinkles are being championed in beauty campaigns, but so are AI-generated images devoid of quirks and diversity.”
They’re good brands – there’s nothing wrong or bad about conglomerates, either. But the list is not surprising. Big brands have big budgets, and the internet has been a ‘pay to play’ space for years. You cannot, as it stands, directly pay an A.I. platform to be prioritised within its conversational answers. But you can pay Google, Meta, publishers and social media influencers, who are the information wells these algorithms draw from. The issue with the A.I. of it all comes from the lack of nuance, the lack of completeness and as one BOF commenter pointed out, the lack of skin (an eerie metaphor for its inherent lack of humanness).
Would you buy directly from someone – something – that’s never once touched, seen or experienced skin in any capacity? It sounds bizarre, but A.I. beauty routines are on the rise. And I get it. Why would we pour over reviews, Reddit, TikTok and Instagram searching for the best products? Why would we pay an expensive dermatologist? A.I. can serve as a research tool, and if you are utterly confused, it will probably give you a good starting point. But the complete lack of nuance is a real problem.
Raquel Medina-Cleghorn, an aesthetician based in N.Y.C., sees its impacts in her clinic all the time: “This sounds basic, but the reason I don’t like A.I. for skincare routine curation is literally because it is not human,” she tells me. “It’s aggregating information from the entire internet, which we know to be rife with misinformation, and spitting it back out at you, sometimes even doing that incorrectly. There is way too much nuance in skincare for blanket recommendations. I just asked A.I. what my routine should be (I’m 40, female, melasma) and it missed some extremely important caveats about one of the ingredients it recommended. It’s a safety thing. Now, if someone is truly novice and has no idea where to start, it’s fine, but I would so much rather hear that someone is following recommendations from a friend or from a Reddit sub because those at least come with experience and human interaction.”
Not only does A.I. possess the power to perpetuate reductive skincare routines, but also the deeper stories around beauty that are being told. As the internet gets more and more centralised, we lose out on information that speaks to different skin types, skin colours, hair textures. If we’re relying on an unregulated, unmonitored information aggregator and not bothering to look beyond, things become worryingly narrow (i.e. in line with ‘typical’ beauty standards pertaining to whiteness, hair texture, thinness) and as a result, homogenised (that dystopic monoculture discussed previously). Beauty has always had a diversity problem – we need to continuously hear from a broad spectrum of founders, experts, creatives, writers and commentators from all cultural backgrounds. It’s absolutely essential.
“Beauty should be a means to experiment, have fun, create, communicate without words. At the crux of it, beauty is inherently emotional and – not to be glaringly obvious – it’s human.”
The same can be said when we use A.I. to search for brands and products that align with our values. There was a secondary report in BOF about factually incorrect information relating to sustainable beauty brands recommended by Claude, ChatGPT and Gemini. Further proof that critical analysis – or at the very least, a grain of salt – is vital.
A.I. also carries profound impacts when it comes to creativity within the beauty industry. A.R.-based video content creation, A.I. models, A.I. generated campaigns, A.I. makeup on either real models or fake ones… These concepts undermine the beating heart of the beauty industry. Image making is an art and creative endeavour that requires taste, passion, and collaboration. Makeup, hair and nail artistry are true vocations that take years of training and dedication to perfect, not to mention meticulous technical skill. Beauty is about vanity in a way, but it’s also about self-expression. This filters down from the industry – those behind the scenes – to consumers. Beauty should be a means to experiment, have fun, create, communicate without words. At the crux of it, beauty is inherently emotional and – not to be glaringly obvious – it’s human.
We are simultaneously living in a time where acne, stray hairs, crooked teeth and folded wrinkles are being championed in beauty campaigns, but so are A.I.-generated images devoid of quirks and diversity. The concept of non-human beauty icons worries me. Interestingly though, for all the hundreds of interviews I’ve done over the years, not a single person has cited something or someone artificial when asked to name a muse, or a hero. In fact, it’s usually a Mother, a Grandmother, or a mentor.
But constantly absorbing all of this perfection is a slippery slope. I recently read about a phenomenon called ‘filter creep’ where A.I. filters are slowly infiltrating every single instance of where we see ourselves online. There’s a similar phrase, ‘perception drift’, that relates to the gradual shift of perception that can accompany ongoing cosmetic procedures. Many of us look at ourselves online more than we look at ourselves in real life; the constantly smoothed out, beautified iterations become ‘real’. Not only this, but many A.I. editing tools are known to dilute or flatten cultural features because models are largely trained on western, Euro-centric date sets. This is referred to as algorithmic bias.
One byproduct of all this is that some brands have wholly embraced imperfections, emotion, community, and nostalgia. In-person pop-ups have more traction than ever as customers crave offline, in-person experiences. People want to touch and feel (beauty has always been tactile, after all). Merit is one example of a brand using nostalgia to speak to its community. Print formats, celebrity faces from the pre-social media era, like Martha Stewart and Lauren Graham. Other examples of those dedicated to real people, places and things include farm-based Flamingo Estate (founded by Australian Richard Christiansen) and Saie Beauty, who host run clubs for brand fans for no other reason than to facilitate IRL connection. It’s my instinct that the brands that will last, or at least deeply resonate, are those that believe in the analogue, believe in experience, and understand that what matters most is the integrity of the process.
Of course, there are huge opportunities at the intersection of A.I. and beauty, as well. It can process complex data, extrapolate key information and trends, assist with budgeting, forecasting, and other administrative duties. This is the hope for most industries, I guess – A.I. can free up our time and bandwidth to focus on more fulfilling, creative or meaningful pursuits. We don’t want it to replace, but to streamline and enhance. Whether we can have one and not the other is up for debate. Time will tell.
Fundamentally, it is important to stay close to real life, especially within beauty. The way we see, express, and care for ourselves is one of the most intimate, individual things we can do. Critical thinking, doubt, and mistakes propel the industry forward, both in terms of creativity and innovation. It takes contrast to create meaning. Different voices, faces, opinions, and perspectives make things better. We really should protect this at all costs
