Must retailers disclose how their recommendation engines work?

Yes. The EU Digital Services Act Article 27 requires online platforms to describe the main parameters used in recommender systems in plain, intelligible language, and to offer users at least one option for recommendations not based on profiling.

Is personalized pricing legal in the EU?

Personalized pricing is not prohibited outright, but the Omnibus Directive and Unfair Commercial Practices Directive require clear disclosure when prices are personalized. Undisclosed personalized pricing may constitute a misleading commercial practice. Several member states have imposed additional transparency requirements.

Does virtual try-on technology involve biometric data under GDPR?

Yes. AI systems analyzing body measurements, facial structure, or skin characteristics process biometric data as defined in GDPR Article 4(14). If used for identification purposes, explicit consent under Article 9(2)(a) is required. A DPIA is advisable in all cases.

How does sustainability regulation affect retail AI?

The proposed Ecodesign for Sustainable Products Regulation requires Digital Product Passports for textiles from approximately 2027. AI systems managing inventory, production, and supply chains must be designed to collect, maintain, and provide the sustainability data required for these passports.

Must AI chatbots identify themselves as AI to customers?

Yes. EU AI Act Article 50(1) requires that persons interacting with an AI system be informed they are doing so, unless this is obvious from the circumstances. Retail chatbots must clearly disclose their AI nature at the start of interaction.

Quick answer

Retail AI must comply with the EU Digital Services Act disclosure requirements for recommender systems (Article 27), the Consumer Rights Directive for pricing transparency, GDPR for profiling and personalization, and emerging sustainability disclosure rules for AI-driven supply chain decisions.

Updated June 2026 · MmowW AI Compliance

AI Compliance in Fashion and Retail: Recommendation Engines and Consumer Rights

Retail AI Regulatory Landscape

Fashion and retail AI applications face a convergence of consumer protection, data protection, and digital services regulation. The primary compliance obligations arise from the EU Digital Services Act (DSA), the Consumer Rights Directive (2011/83/EU), the Unfair Commercial Practices Directive (2005/29/EC), GDPR, and emerging sustainability regulations. The EU AI Act adds requirements for AI systems making decisions that significantly affect consumers.

Regulatory Requirements by AI Application

Retail AI Application	Primary Regulation	Key Obligation
Product recommendation engines	DSA Article 27, GDPR Article 22	Disclose main parameters of recommender system; right to non-profiled recommendations
Dynamic pricing	Consumer Rights Directive Art. 6a (Omnibus Directive amendment)	Show prior price when advertising reductions; no personalized pricing deception
Virtual try-on / AR fitting	GDPR Articles 9, 35; EU AI Act Article 50	Biometric data protection; AI-generated content disclosure
Inventory and demand prediction	General commercial law; sustainability reporting	Anti-waste obligations; supply chain due diligence
Customer service chatbots	EU AI Act Article 50(1); Consumer Rights Directive	Disclose AI interaction; maintain consumer rights to human agent
Size recommendation AI	GDPR Article 22; product liability law	Accuracy obligations; return cost implications

Recommender System Transparency

The Digital Services Act Article 27 requires online platforms to disclose the main parameters used in their recommender systems in plain language. For fashion retail, this means explaining how product recommendations are generated: whether based on purchase history, browsing behavior, demographic data, trending products, or commercial partnerships. Users must be offered at least one recommendation option not based on profiling.

GDPR Article 22 grants individuals the right not to be subject to decisions based solely on automated processing, including profiling, that produce legal or similarly significant effects. A recommendation engine that determines credit eligibility, insurance pricing, or access to services triggers this right. Product recommendations alone generally do not meet the significance threshold, but personalized pricing that materially affects what a consumer pays may qualify.

Dynamic Pricing Compliance

The Omnibus Directive (2019/2161) amended the Consumer Rights Directive to require retailers to display the lowest price from the prior 30 days when advertising price reductions. This directly constrains AI-driven dynamic pricing. When an AI adjusts prices based on demand, competitor pricing, or inventory levels, the historical price disclosure requirement applies to any advertised reduction.

Personalized pricing based on individual consumer profiles raises additional concerns under the Unfair Commercial Practices Directive 2005/29/EC. The European Commission guidance specifies that personalizing prices without clear disclosure is potentially misleading. German courts have found undisclosed personalized pricing to violate the UWG (Gesetz gegen den unlauteren Wettbewerb). France's Loi Hamon requires explicit disclosure when prices are personalized.

Biometric Data in Virtual Try-On

AI-powered virtual try-on features that analyze body measurements, facial features, or skin tone process biometric data as defined under GDPR Article 4(14). When used for identification purposes, this constitutes special category data under Article 9, requiring explicit consent. Even when not used for identification, a DPIA under Article 35 is advisable given the sensitivity of body measurement data.

The EU AI Act Article 50 requires disclosure when consumers interact with AI-generated content, including AI-altered images showing products on virtual models of the consumer.

Sustainability and AI-Driven Supply Chains

The EU Strategy for Sustainable and Circular Textiles and the proposed Ecodesign for Sustainable Products Regulation (ESPR) create obligations that affect AI-driven inventory and production decisions. AI systems optimizing fast fashion production cycles must account for extended producer responsibility requirements, textile waste reduction targets, and the Digital Product Passport requirement (expected from 2027 for textiles) which mandates detailed supply chain information for each product.

The Corporate Sustainability Due Diligence Directive (CSDDD) requires large fashion companies to identify and address adverse human rights and environmental impacts in their value chains. AI supply chain optimization systems must be designed to support rather than undermine these due diligence obligations.

Consumer Rights in AI Interactions

When AI chatbots handle customer service, the Consumer Rights Directive right of withdrawal (14-day cooling-off for online purchases) must be clearly communicated by the AI system. AI systems processing returns or complaints must not create barriers that a human agent would not impose. Under EU AI Act Article 50(1), consumers must be informed when they are interacting with an AI system.

Practical Compliance Steps

Publish a clear description of recommender system parameters as required by DSA Article 27
Implement 30-day prior price tracking for any AI-driven price reduction advertising
Obtain explicit consent before processing biometric data for virtual try-on features
Disclose AI interaction clearly in chatbot interfaces per EU AI Act Article 50
Ensure AI-optimized supply chains support Digital Product Passport and due diligence requirements
Offer consumers a non-profiled recommendation option alongside personalized suggestions

Check your AI compliance readiness — free.

Take the Readiness Check 3 minutes · 10 questions · no signup required

This article is for informational purposes only and does not constitute legal advice. Regulatory requirements change frequently — verify current rules with official sources. Built by Sawai Gyoseishoshi Office, Hiroshima, Japan.