The Cut Comes Before the Cloth: What London Fashion Week Reveals About Design in the Age of AI Images

An AI can draw a dress in seconds. It still can't tell you how to build one — and that gap is where fashion design actually lives.

On 17 September 2026, the industry gathers in London for London Fashion Week SS27, running to the 21st, with Alexander McQueen returning to the schedule — the house noting that "London is where our story began." It arrives at a strange moment for fashion. Generative AI can now produce a photorealistic garment in seconds; the runway has never had more competition from the feed. And yet the clothes on the LFW catwalk make a quiet, expensive point: an image of a dress is not a dress.

The gap is technical, and it is the whole game. As one industry analysis puts it, generative AI produces "raster images — pixels without pattern data, manufacturing specifications, or sustainability metrics" — you cannot take a rendered image and send it to a factory. A picture shows a result. It says nothing about the pattern that has to lie flat on a cutting table, the seam that has to carry load, or the way cloth falls when a body moves inside it.

That translation — from image to buildable garment — is fashion's real concept phase, and it happens long before the show, in the pattern and the toile: a garment mocked up in cheap calico so the three-dimensional truth of a cut can be tested before a metre of real fabric is touched. London is a fitting place to make the point, because its design culture is unusually construction-led — its schools and ateliers still treat cut and pattern as the substance of design, not its decoration.

The industry is not ignoring AI — it is racing toward it. Roughly 40% of apparel prototyping now involves AI assistance, and a similar share of brands have folded AI pattern-making into at least one stage of design. The serious tools reflect exactly that lesson — they don't stop at a pretty picture. Platforms now turn a sketch into a true 3D garment with pattern data and fabric simulation, generate sewing patterns directly from images, and researchers are building physics-aware representations so a design and its manufacturable pattern stay in sync. The direction of travel is telling: the value isn't generating more images, it's generating things that can actually be made — a shift echoed across academic work on AI pattern generation.

This is the concept-phase discipline in its purest form. The most consequential decision in a collection — the silhouette, the cut, the way a shape sits on the body — is made upstream, in three dimensions and real material, and everything downstream serves it. A house with a strong concept phase can use AI to explore a hundred cuts before lunch. A house without one just generates a hundred images it can't build. As guides to AI fashion design keep repeating, the tool amplifies the process you already have; it does not supply one.

That is why LFW still matters in an age of infinite fashion imagery. The published schedule is not a gallery of pictures — it is a public proof that these clothes resolve in reality: cut, sewn, fitted and made to move. The official platform exists to certify the one thing an image never can — that the design survived contact with construction.

The lesson travels well beyond fashion. Every field now has tools that generate a beautiful result and skip the hard question of whether it can be built. The enduring advantage is a concept phase that already reasons in three dimensions and real constraints — that treats an image as a question, not an answer. Deciding a form so it carries all the way through to something you can actually make is exactly the work we obsess over at Depix.

Sources:

• ●London Fashion Week (British Fashion Council)
• ●London Fashion Week — Sep 17-21, 2026 (World Footwear)
• ●McQueen to return to London Fashion Week SS27 (Luxury London)
• ●Turning fashion sketches into accurate 3D models (Style3D)
• ●The digital pattern-making revolution: why 2025 changes everything (FashionINSTA)
• ●Can AI make sewing patterns effectively? (Style3D)
• ●AI sewing pattern generators from images (Style3D)
• ●Textile IR: physics-aware intermediate representation for fashion CAD (arXiv)
• ●AI-assisted Pattern Generator for Garment Design (ResearchGate)
• ●What is AI Fashion Design? Complete Guide 2025 (WearView)
• ●London Fashion Week Schedule (Fashion Week Online)
• ●London Fashion Week — About