How to brief a board on AI: a guide for fashion executives

Boards are under pressure: 66 percent of directors already use AI for board work but only 22 percent of boards have a formal AI usage policy, and only about 15 percent of boards currently receive AI-related metrics. Fashion executives preparing briefings need to address three recurring questions: who owns the risk, where is the money going, and how do we measure value. Recent guidance from director associations and governance advisers shows how to structure answers that satisfy fiduciary expectations without requiring technical expertise.

What questions do boards ask about AI risk?

Every major board governance survey in the past 12 months identifies AI as the number one strategic risk that boards feel least prepared to oversee. In 2026, advisers describe AI governance as a continuous, board-wide obligation rather than a delegable, episodic check-in, comparable to how cybersecurity oversight evolved a decade earlier.

The National Association of Corporate Directors outlines four steps that every board should consider in the AI age: conduct a comprehensive assessment of AI use and risks across the organisation; establish effective oversight structures for AI governance; implement protocols for identifying and managing AI-related risks aligned with recognised frameworks; and empower teams to proactively leverage AI opportunities.

Boards typically probe three risk categories. The guidance addresses how AI reshapes corporate risk profiles, including exposure to hallucinations and algorithmic bias. In fashion specifically, legal counsel flag AI-generated output requiring human review to prevent hallucinations in product descriptions, sizing guidance or customer service; use of AI models and influencers triggering right of publicity considerations requiring written consent; data and privacy concerns where personalisation features and virtual try-ons may involve consumer data including biometric identifiers; and vendor due diligence addressing training data provenance, permissible commercial uses, security requirements and indemnities for intellectual property claims.

Who should own AI oversight at board level?

Put AI on the standing board agenda, assign clear oversight ownership to audit, risk, or a dedicated technology committee, and require management to report AI inventory, incidents, and usage-policy compliance at every meeting. Only about 39 percent of Fortune 100 boards have explicit AI oversight mechanisms such as a board committee, a director with AI expertise, or a dedicated governance sub-board, according to McKinsey's research on how boards can evolve their AI governance.

Best practice varies by risk profile: a dedicated AI and Technology Committee is suited for organisations where AI is central to the business model, holding primary oversight responsibility for AI strategy, risk and ethics with a mandate that includes reviewing material AI deployments and receiving regular updates from the Chief AI Officer or equivalent; an Audit and Risk Committee expansion is practical where AI risk is primarily a compliance and control concern; full board ownership is suitable for smaller boards.

Only 28 percent of organisations put the CEO directly in charge of AI governance oversight and only 17 percent give that role to the board, which means accountability is still diffuse at the exact moment regulators are removing the ambiguity from that question.

How much are fashion companies spending on AI?

Major global fashion players plan to channel upwards of 40.9 billion dollars into investments by 2026, prioritising spending on AI, storefront innovation, omnichannel delivery and supply chain optimisation. A total of seventeen fashion companies plan to increase their combined investment this year to approximately 40.9 billion dollars, compared to the nearly 36.3 billion spent in 2025. Individual company increases are substantial: Capri Holdings is increasing its investment in AI by 98.4 percent, from 63 million dollars in 2025 to 125 million.

The AI in fashion market will grow from 1.75 billion dollars in 2025 to 2.47 billion dollars in 2026 at a compound annual growth rate of 40.8 percent. McKinsey estimates that generative AI alone could add 150 billion to 275 billion dollars to apparel, fashion and luxury operating profits within three to five years.

While 92 percent of companies say they will increase their investments in generative AI, only 1 percent say their deployment of AI has reached maturity. Many are stuck in pilot mode, testing siloed AI solutions with limited impact, with scaling often seen as too complex or costly. There is no way to meet with the board of directors without talking about AI, though the big challenge is developing a plan that makes the return on investment tangible.

Which AI metrics should boards track?

Only about 15 percent of boards currently receive AI-related metrics. Boards should have access to impact measures such as ROI by business unit, percentage of processes that are AI enabled, resilience indicators such as override rates and backup drill results, workforce-reskilling progress, and regulatory alignment. Directors should inquire about the specific KPIs being used to measure AI performance, bias and accuracy. It is no longer sufficient to measure AI success solely through the lens of cost savings or speed. Governance requires looking at the cost of error and the frequency of intervention.

Five core metrics recur in governance frameworks:

• AI inventory coverage: percentage of AI systems registered and classified by risk level, including third-party tools
• Control maturity: number of high-risk systems meeting defined governance standards for human oversight, explainability and auditability
• Incident response readiness: escalation triggers, particularly for customer-facing AI, regulated data or defined exposure thresholds
• Regulatory compliance status: tracking adherence to frameworks such as the EU AI Act, NIST AI Risk Management Framework and sector-specific rules
• Third-party AI risk coverage: vendor assessment addressing training data provenance, security requirements and indemnities

Boards must ensure management maintains a real-time inventory of all AI systems deployed within the organisation, including third-party tools. Boards should mandate that management establishes a centralised AI inventory. Without knowing exactly which models are running across various business units and who is responsible for their maintenance, the board cannot effectively oversee the risk profile. Metrics must be integrated into regular board reporting cycles, ensuring that AI risk is treated with the same severity as financial or operational risks.

What regulatory obligations apply in 2026?

The EU AI Act's general rules took effect 2 August 2026. The EU AI Act, adopted in July 2024 and entering phased effect beginning in 2025, establishes binding obligations for organisations deploying AI in European markets, including conformity assessments, human oversight and documentation standards for high-risk systems under a formal AI governance framework. The NIST AI Risk Management Framework has become the de facto voluntary standard in the United States.

Directors in jurisdictions with fiduciary AI obligations now face personal liability exposure if they cannot demonstrate that their organisations have adequate oversight mechanisms in place. The EU AI Act, the UK AI Security Institute guidance and SEC cyber disclosure rules all converge on a single expectation: boards must demonstrate active oversight of material AI risks, not just receive management reports.

The EU AI Act and various US state-level laws are pushing toward broader AI imagery disclosure requirements. By 2027, fashion brands using AI-generated imagery in advertising may face mandatory labelling in multiple major markets. Brands that have built disclosure into their AI workflows now are positioned ahead of compliance; brands that have not are positioned to be reactive.

How should executives structure a board AI briefing?

Governance guidance converges on a four-part structure. Begin with posture: Most organisations still lack a clear view of how AI fits into their strategy or transformation agenda. Without alignment between the board and management, oversight becomes either superficial or paralysing. Boardroom test questions include: which competitive advantages does AI enhance or threaten; does our AI business case target a large enough value pool to reshape our market; what key resources are needed to do that and do we currently have them; do we have dedicated people who can manage the environmental, regulatory, legal and reputational risks that come with being a business pioneer; and how do we need to evolve our innovation pipeline to match the pace of technological change.

Second, present the inventory. Enterprise-level fashion buyers now need to ask: who owns the data; is customer content used for model training; where is the data processed; can outputs be reviewed, traced and audited; and can the workflow support legal, IT and procurement review.

Third, demonstrate controls. A core requirement of both the EU AI Act and the NIST AI RMF is ensuring meaningful human intervention authority over AI-driven decisions. Boards must verify that human-in-the-loop processes are operational, not just documented, particularly for high-risk systems that affect employment, lending or healthcare decisions. A 2025 analysis found that 22 percent of companies consider AI inaccuracies or bias as a material risk. Directors have a responsibility to not only ask how models are validated and what data they are trained on but also to verify data themselves and take accountability for any decisions made with the aid of AI.

Fourth, define value capture. At Zalando, generative AI has reduced image production costs by 90 percent. According to SimilarWeb, ChatGPT accounted for 16 percent of Zara's and 8 percent of H&M and Aritzia's inbound traffic between June and August 2025. 53 percent of US consumers who used generative AI for search in Q2 2025 also used it to help them shop. Unlike paid ads which are clearly sponsored, over 40 percent of consumers say AI responses feel more reliable, while only 15 percent consider them less reliable. Shoppers are therefore increasingly willing to make purchases from AI-based recommendations, as seen in the 84 percent growth in AI-driven revenue per visit on US retail sites between January and July 2025.

McKinsey finds that fewer than 25 percent of companies have board-approved AI governance policies, and only 15 percent of boards receive metrics like ROI, override rates, explainability indicators or reskilling progress. Executives who provide this structure position AI as a governed strategic capability rather than an unmanaged technology experiment.

Researched and drafted with AI support, reviewed and released by the editorial team.

EN DE FR IT