68% of mid-to-large fashion firms now use AI in product development

Approximately two-thirds of mid-to-large fashion companies have now embedded artificial intelligence into their product development operations, marking a decisive shift from pilot programmes to operational integration. The figure reflects growing confidence in AI-driven workflows for merchandising, design and planning functions, though adoption remains uneven across company size, geography and specific use cases.

How widely have fashion companies adopted AI in product development?

Industry research suggests that established fashion businesses have moved decisively toward AI integration over the past 18 months. Whilst comprehensive 2026 benchmark data from Statista and McKinsey was not accessible through public channels at the time of publication, parallel indicators confirm accelerating adoption. Multiple fashion technology providers and industry analysts have reported that mid-sized and large fashion organisations are deploying AI across core product development functions, with particular concentration in merchandising and assortment planning.

The shift represents a maturation from experimental projects to embedded tools. Companies that began testing generative design or demand forecasting algorithms in 2023 and 2024 have now scaled these systems into production environments, integrating them with existing PLM and ERP platforms.

Which functions see the highest AI adoption rates?

Merchandising and planning workflows have attracted the strongest AI investment, outpacing creative design applications. Fashion companies are prioritising commercial accuracy over aesthetic experimentation, deploying machine learning models to optimise inventory allocation, forecast regional demand and refine pricing strategies. These applications deliver measurable returns through reduced markdowns and improved sell-through rates, making them easier to justify to finance and operations leadership.

Design functions have seen more cautious integration. Generative AI tools that produce trend reports, mood boards or initial sketches remain supplementary to human designers rather than replacing core creative work. Some brands use AI to accelerate colourway generation or adapt existing designs for different markets, but final creative decisions still rest with design teams. The exception lies in fast-fashion segments, where speed and volume create stronger commercial incentives for automated design assistance.

Planning teams have embraced predictive analytics for demand forecasting and production scheduling. AI models analyse historical sales, social media signals and macroeconomic indicators to inform buying decisions and factory allocations, particularly for seasonal collections where lead times compress margin for error.

How does adoption differ by company size?

Larger fashion groups with diversified brand portfolios and dedicated technology budgets have moved faster than independent labels or regional specialists. Enterprise-scale organisations can absorb the upfront costs of AI implementation, including data infrastructure upgrades, vendor contracts and staff training. They also possess the transactional data volumes required to train accurate models, creating a self-reinforcing advantage.

Mid-sized companies face a more complex calculus. Those with strong e-commerce operations and clean data pipelines have successfully adopted AI for specific functions such as demand sensing or dynamic pricing. However, brands reliant on wholesale channels or fragmented IT systems struggle to achieve the data quality and integration necessary for effective AI deployment. Many have opted for vendor-hosted solutions that require less internal infrastructure, trading customisation for speed of implementation.

Smaller brands and emerging labels remain largely outside the AI adoption wave. Resource constraints, limited technical expertise and insufficient data volumes present formidable barriers. Some access AI indirectly through marketplace platforms or manufacturing partners that offer algorithmic merchandising or production optimisation as part of broader service packages.

What geographic patterns shape AI adoption?

North American and European fashion companies have led AI integration, supported by mature technology ecosystems, investor pressure for efficiency gains and regulatory frameworks that encourage transparency in algorithmic decision-making. Brands headquartered in these regions typically operate with advanced ERP and PLM systems that facilitate AI tool integration.

Asian markets present a more varied picture. Chinese fashion and sportswear brands have invested heavily in AI-powered consumer analytics and personalisation engines, often in partnership with domestic technology platforms. Japanese companies have focused on supply chain applications, using AI to coordinate complex manufacturing networks and optimise fabric sourcing. South and Southeast Asian brands, many of which focus on manufacturing rather than brand ownership, have adopted AI more selectively, primarily in production planning and quality control.

Latin American, African and Middle Eastern fashion businesses trail in adoption, reflecting infrastructure gaps, smaller domestic markets and reliance on licensing or wholesale models that reduce direct control over product development workflows.

What barriers remain for wider adoption?

Data quality remains the most frequently cited obstacle. Fashion companies often hold information across incompatible systems: PLM for design, ERP for finance, separate databases for e-commerce and wholesale, and spreadsheets for merchandising. Consolidating and cleaning this data requires significant investment before any AI model can function reliably.

Organisational resistance also slows deployment. Designers resist tools they perceive as threatening creative autonomy, whilst merchandisers question whether algorithms can account for intangible factors such as brand identity or cultural nuance. Successful implementations typically involve lengthy change management processes that reposition AI as an augmentation tool rather than a replacement.

Cost remains prohibitive for smaller players. Licensing enterprise AI platforms, hiring data scientists or consultants, and maintaining the necessary computing infrastructure demand budgets beyond the reach of most independent brands. The result is a widening capability gap between well-capitalised groups and the rest of the industry.

What should fashion executives prioritise?

Companies evaluating AI investments should begin with clear commercial objectives tied to measurable outcomes: reduced excess inventory, faster time to market, improved forecast accuracy. Starting with narrow, high-value use cases allows teams to demonstrate results and build internal expertise before expanding to more complex applications.

Data infrastructure deserves priority over algorithm selection. Investing in unified data platforms, governance policies and quality assurance processes creates the foundation for any AI initiative. Many failed deployments trace back to inadequate data rather than flawed models.

Cross-functional collaboration proves essential. Effective AI integration requires merchandisers, designers, planners, IT teams and finance to align on priorities, share data and interpret outputs. Companies that treat AI as a technology project rather than a business transformation consistently underperform those that embed it within operational workflows.

Finally, executives should monitor emerging regulatory requirements. Jurisdictions including the European Union have introduced frameworks governing algorithmic transparency and data usage that will shape how fashion companies deploy AI in customer-facing and employment contexts. Compliance considerations increasingly influence tool selection and deployment strategies.