AI budget planning for fashion: how to split spend on licences, data and people

How much are fashion companies spending on AI in 2026?

Seventeen fashion companies plan to invest approximately 40.9 billion dollars in technology and AI in 2026, compared to nearly 36.3 billion dollars spent in 2025, according to analysis by Standard & Poor's reported by Modaes. Capri Holdings is increasing its investment from 63 million dollars in 2025 to 125 million dollars in the current fiscal year, a 98.4 per cent rise.

The fashion industry's appetite for AI investment is clear. Whilst 92 per cent of companies say they will increase their investments in generative AI, only 1 per cent say their deployment of AI has reached maturity, according to the BoF-McKinsey State of Fashion 2026 Executive Survey. Most remain stuck in pilot mode, testing isolated solutions with limited impact.

Fashion sits within a broader retail transformation. Walmart plans to spend 25 billion to 27 billion dollars on capital expenditure this year, whilst total expenditures by the other 16 retailers in a study are seen jumping 25 per cent to 13.9 billion dollars, at the midpoint of 2026 estimates, according to Women's Wear Daily.

How should fashion budgets split between software, data and people?

Finance leaders outside fashion offer clearer guidance than the industry itself. CFO guidance suggests four pillars for AI budgets: data and infrastructure at 20 to 30 per cent, AI use cases and platforms at 30 to 40 per cent, risk and governance at 15 to 25 per cent, and people and change at 15 to 25 per cent, according to finance technology publication Zenskar.

Enterprise allocations confirm this pattern. Enterprise AI budgets in 2026 are allocated roughly 35 per cent to software and SaaS AI tools, 22 per cent to cloud infrastructure, 17 per cent to internal AI talent, and the remainder split across implementation, data platforms and governance, according to venture capital analysis published in July 2026.

A narrower framework also circulates. One CFO guide proposes a 40-30-20-10 allocation model: 40 per cent infrastructure and data foundations, 30 per cent talent and capability building, 20 per cent governance and compliance, 10 per cent innovation and experimentation. The precise split varies by company size, maturity and regulatory exposure, but the underlying principle holds: software and infrastructure claim the largest shares, followed by talent, with governance taking a growing slice.

Yet a warning emerges from the data. Deloitte's State of AI 2026 found that 93 per cent of AI budgets go to technology, tools, licences and compute, and only 7 per cent toward the people and workflows expected to drive value from those tools. That imbalance may explain why many pilots fail to scale.

What are CFOs prioritising in 2026 AI budgets?

Technology budgets are set to rise for 75 per cent of CFOs, with nearly half planning increases of 10 per cent or more. Nearly 60 per cent of CFOs plan to increase finance function AI investments by 10 per cent or more in 2026, whilst another 24 per cent expect gains of 4 to 9 per cent, according to Gartner research published in February 2026.

Sixty one per cent of CFOs say enterprise AI spend will increase by 5 to 20 per cent in 2026, whilst 16 per cent see faster growth, according to the Oliver Wyman Forum's CFO Agenda survey. The shift from cautious pilots to committed scale is driven by tangible gains in productivity and decision making, though measurement remains uneven.

CFO AI budgets are shifting from pilot experimentation toward structured deployment and measurable ROI. In 2026, AI spending is moving into operational technology budgets with the same rigour applied to ERP investments or headcount decisions, according to finance AI publication ChatFin.

Where does infrastructure spending actually go?

Cloud and compute dominate infrastructure budgets. Major cloud providers cut prices on high-end AI hardware by roughly 40 to 45 per cent in mid-2025 as next-generation chips expanded supply. On-demand rates for top-tier AI hardware now run approximately 3 to 4 dollars per GPU-hour. Longer-term commitments bring effective rates below 2 dollars per GPU-hour, according to cloud cost management platform CloudZero.

Organisations are spending 40 to 60 per cent more on AI infrastructure than they budgeted, driven by unexpected compute costs, data storage requirements and the need for redundancy and reliability at scale, according to strategy consultancy Beehive. For a Fortune 500 company running production AI systems, monthly infrastructure costs can easily exceed 2 to 5 million dollars.

Data preparation carries hidden costs. Fashion companies cannot simply buy software and expect results. AI systems require clean, structured data, and most fashion organisations underestimate the effort required to prepare legacy inventory, sales and customer data for AI consumption. Infrastructure budgets must account for data engineering, storage and governance, not merely model hosting.

What skills should fashion companies hire or train?

AI expertise, especially at executive level, will need to be seriously considered to steer organisations through developing the product roadmap, ethical considerations and value creation, according to fashion recruitment consultancy Virua quoted by Drapers in early 2025.

Recent industry analyses, including the BoF-McKinsey State of Fashion 2026 report and recruiter insights, show that low single-digit growth, tariff pressures and rapid AI adoption will prioritise hybrid skills that blend creativity with business and tech acumen. Understanding AI tools for design, trend forecasting, customer service, content creation and supply chain optimisation is the top opportunity cited by executives, according to Digital Fashion Academy.

Employers now prioritise skills in data analysis, AI integration and digital tools alongside traditional merchandising expertise to keep pace with rapidly evolving fashion retail demands. Recent reports highlight a 35 per cent surge in demand for AI-related skills within retail and fashion over the past three years, according to education research platform Research.com.

Fashion businesses face a build versus buy decision not only for software but also for talent. Some retailers, including Levi Strauss, have run internal AI boot camps since 2021 to upskill existing employees. Others hire externally. The 15 to 20 per cent talent allocation in enterprise AI budgets includes both new hires and training programmes for current staff.

When should fashion companies fund AI initiatives?

Timing matters as much as allocation. AI's share of the IT budget has more than tripled, from 5 to 17 per cent. The survey also suggests that AI spending outside the Tech/IT budget is rising significantly as companies move from pilots to enterprise-wide programmes, according to BCG's Applied AI 2026 survey.

Fashion companies should prioritise high-ROI use cases first. Demand forecasting, inventory optimisation and size recommendation systems deliver measurable returns within months. AI-driven demand forecasting cuts forecast errors by 20 to 50 per cent and can lower inventories by a similar range whilst reducing lost sales by up to 65 per cent, according to industry consultant Tommaso Maria Ricci. Marketing and design applications follow, with infrastructure and governance running in parallel from the start.

Budget cycles should account for iteration. Unlike traditional IT projects with fixed scopes, AI initiatives require ongoing refinement as models retrain and use cases expand. Finance teams should structure AI budgets with quarterly review gates, not annual approvals, to allow reallocation toward initiatives that demonstrate traction and away from those that stall.