When it comes to Governance, Retailers need a control plane for context

Summary

Retailers are moving from isolated generative-AI pilots to enterprise-wide adoption, where each store, headquarters, and digital workflow needs different data, permissions, models, and cost controls. The post argues that useful AI depends on governed business context, while fragmented tools can produce separate provider contracts, logs, permission models, cost centers, and answers to the same question. It proposes an open control plane that centralizes governance and model access without forcing employees into one application or vendor. The described Databricks design combines Unity Catalog for data and business context, Unity AI Gateway for routing and oversight, Foundation Model APIs for commercial, open-source, and custom models, and Genie for natural-language queries over governed data; the intended result is faster experimentation and deployment with visibility into access, spending, lineage, and auditability.

Context

Retailers are shifting from isolated AI pilots to enterprise-wide adoption across stores, headquarters teams, and digital channels. The source identifies fragmentation as the central challenge: different tools and teams may maintain separate model access, data access, logs, permissions, costs, contracts, and business logic. It also warns that a disconnected AI layer could duplicate governance work and separate AI systems from existing governed data estates.

Approach / What changed

The source proposes an open AI control plane for context. Unity Catalog governs data, permissions, lineage, models, and trusted definitions; Unity AI Gateway governs model routing, usage visibility, logging, budgets, and rate limits; Foundation Model APIs provide centralized access to commercial, open-source, and custom models; and Genie supports natural-language questions over governed data. Together, these capabilities provide a shared governance and model-access strategy across applications, agents, and interfaces.

Takeaways

• Enterprise AI needs role-specific access to data, models, tools, and permissions rather than one model or one assistant for the whole retailer.
• Fragmented AI tools can create separate provider contracts, prompt and response logs, cost centers, permission models, and conflicting answers to business questions.
• An open control plane is intended to preserve model flexibility while giving IT visibility into data access, model usage, workflow costs, lineage, and auditability.

Related reading