Grit and Glitch: When AI Meets the Fashion Editorial
I wanted to test AI, specifically Sora, and its ability to mimic the high drama of a gritty, grungy fashion shoot. Not a polished beauty spread or some glossy corporate ad, but the kind of shoot that lives in dim light and dirty stairwells. Where the styling is intentional chaos and the whole thing smells like leather and rain.
About twenty prompts later, it delivered.
The results were, unexpectedly, good. Startlingly good, actually. Not just technically competent, but evocative. The images carried the right mood. The models had the too-cool detachment. The shadows felt heavy. You could almost hear the soundtrack: industrial, with a hint of 3am melancholy. If I hadn’t told you it was AI-generated, you might’ve scrolled right past thinking it came from a half-forgotten i-D editorial or some underground zine with a cult following.
Sure, there were glitches. Still a bit of that uncanny shimmer around hands, fabric behaving like it’s never seen a wind tunnel, expressions that flicker between bold and blank. But you had to look for it. And the thing is, those glitches are disappearing. Fast. Give it a year, maybe less, and you’ll be able to summon an entire campaign with the right cocktail of moodboard, prompt, and post-work. No jet lag. No day rates. No weather delays.
This leads to the real question, at least the one I’m chewing on now: will people care?
Will it matter if the model was never cold? If the grit came from code, not concrete? If the rain in her hair never touched real sky?
Maya Angelou once said, “People will forget what you said, people will forget what you did, but people will never forget how you made them feel.” Seth Godin reframed it for the branding age: “People do not buy goods and services. They buy relations, stories, and magic.”
And that’s the friction point. Because for all the visual accuracy AI can now conjure, lighting, styling, facial nuance, even camera ‘mistakes’ and what it still struggles to replicate is the mess. The absurd humanity of it all. The bleary-eyed crew setting up gear at 6am. The assistant sprinting down a wet street with the wrong lens. The frozen model pretending it’s August. The creative director’s playlist clashing with the photographer’s nerves. The weird breakfast sandwich that accidentally makes it into a BTS shot and becomes legend.
Is that romantic? Maybe. But it’s also connective tissue. The behind-the-scenes pain, the chaos, the craft, those things wrap around an image like invisible thread. They tell a story before the story. A perfectly generated AI image may stun, but will it stick? Will it live in our heads the way a photo taken on three hours of sleep and a gut feeling sometimes does?
I don’t have a final answer, only a theory. The more AI levels the visual playing field, the more people will crave the unpolished, the process-driven, the raw. Not just the outcome, but the journey. The story of the story. The parts that can’t be replicated because they weren’t planned in the first place.
So maybe the future of creativity isn’t about resisting AI. Maybe it’s about leaning into what AI can’t fake (yet): the quirks, the hesitations, the shot that wasn’t supposed to work but somehow did, the pixel that holds proof of presence, the accident that becomes the idea.
How do we prove that? Still working on it.
Author’s Note:
I started this test partly out of curiosity, partly out of creative anxiety. Could a machine pull off the kind of shoot I used to sweat through in real life? Apparently, yes. But that unsettling success also reminded me of something else: I don’t just miss the outcome, I miss the process. The hauling, the hurrying, the human noise. Maybe what we’ll end up selling in the future isn’t just an image or a product, but the evidence of having been there. And maybe, just maybe, AI’s rise will force us to show our fingerprints a little more boldly. If nothing else, I’ll keep prompting until something breaks, or something beautiful slips through the cracks.
I wanted to test AI, specifically Sora with its ability to create something that looked like a gritty, grungy fashion shoot.
About twenty prompts later, it delivered. The result? Surprisingly convincing. Almost real.
Sure, there’s still a bit of lingering weirdness in the details if you look closely. But those glitches are disappearing faster than you’d expect. And when they’re gone?
It’ll be nearly impossible to tell the difference between a high-budget professional shoot and one conjured from a well-crafted prompt.
Which leads to the question at least the one I’m asking today: will consumers care?
As Maya Angelou once said, “People will forget what you said, people will forget what you did, but people will never forget how you made them feel.” Or, in more marketing-specific terms, Seth Godin nailed it: “People do not buy goods and services. They buy relations, stories, and magic.”
But here’s the dilemma: without the messy, human stuff, the bleary-eyed crew at 6am, the frantic coffee runs, the London-to-Southend gear haul, the half-frozen model in a summer dress—is there still a story? Is there anything left for people to connect with?
My take: as AI levels the visual playing field, the truly standout moments won’t come from flawless prompts. They’ll come from what feels genuinely human. From quirks. From imperfections. From proof of the journey.
How we prove that? I’m still working on it and need a coffee…
Leave a Reply