Research and EU regulation for fashion AI, digested
The Research digest section of Fashion AI Lab EU.
Benchmark figures in fashion AI papers and vendor decks rarely survive close scrutiny. This tutorial gives research, product, and procurement leads a repeatable method for stress-testing any accuracy claim before it enters a decision or a citation.
Read more
The Corporate Sustainability Reporting Directive imposes a structured set of ESRS datapoints on large apparel and footwear groups—but which are mandatory, which depend on materiality, and how far does the reporting boundary extend into outsourced supply chains? This article sets out the answers.
Read more
As EU skills research confirms accelerating replacement demand across craft occupations, fashion organisations face a concrete problem: the tacit judgment of senior patternmakers leaves with them. This article sets out six approaches the industry is using to hold on to it.
Read more
Physics-based and neural learned-drape pipelines each solve garment simulation differently. This comparison helps engineers and product leads choose the right approach for their accuracy, speed, and integration requirements.
Read more
A critical annotation of five widely cited academic papers on AI-driven fashion trend forecasting, examining methodology, data sources, and the gap between reported results and production-ready deployment.
Read more
A catalogue of six open research datasets covering garment images, body measurements, fabric properties, and cloth simulation benchmarks, with notes on licence terms, dataset size, and the model architectures each suits best.
Read more
A curated, annotated list of seven cloth simulation and learned-drape research papers, summarising the method, the results, and what each means for engineers and product managers building or buying fashion technology.
Read more
A new report signals that AI-powered search assistants are fundamentally altering how consumers discover fashion products online, requiring brands to move beyond keyword optimisation toward structured, semantically rich content strategies.
Read more
Industry commentary and research point to a widening gap between fashion brands' stated AI ambitions and their actual governance and data readiness. Technology strategists need to understand where the friction is concentrated.
Read more
Aggregate metrics obscure whether a fashion AI pilot is working. This guide shows technology and business leads how to define KPIs that are specific enough to act on—scoped by operation, category, and time window—drawing on NIST's MEASURE function and RAND's research into AI project failure.
Read more
Gap Inc. established a dedicated Office of AI and a Google Cloud partnership spanning Gemini, Vertex AI and BigQuery. This analysis extracts the organisational lessons for European heads of data governance — while being explicit that Gap's public disclosures address no EU obligation.
Read more
The AI-native versus AI-integrated framing is everywhere in PLM marketing, but no independent literature uses it as an architectural category. This article sets out what each side is actually claiming and gives architects the questions that separate rebuilt from bolted-on.
Read more
Luxurynsight's acquisition of Heuritech consolidates two of the most-cited AI trend intelligence platforms in luxury fashion. Technology and procurement teams need to reassess data access, model independence, and the structural risks that platform roll-ups introduce.
Read more
AI proofs-of-concept in fashion organisations fail for predictable, structural reasons. This article diagnoses the root causes and gives technology leads a framework for deciding whether a stalled pilot is worth rescuing.
Read more
A practical preparation framework for brand product and technology teams attending PI Apparel events, and for solution providers exhibiting or presenting at them.
Read more