Workday, an enterprise cloud applications provider serving more than 10,000 organizations worldwide, needed to scale AI agents across its platform without fragmenting data or eroding trust. As Workday deployed AI agents across departments including procurement, supply chain, finance, and HR, each agent with its own independent data store created point-to-point connections that multiplied complexity exponentially; with 100 agents and 100 data systems, Phoenix Majumder, Senior Director of AI Engineering and Platforms at Workday, described the architecture as "a very complex spaghetti" that was impossible to govern, observe, or monitor with confidence. Majumder warned that at scale, a single wrong decision by an agentic workflow could erode the whole trust plane around AI. Workday therefore made a deliberate architectural decision to convert the multiplicative problem (M × N) into an additive one (M + N), guided by centralization of data, centralization of governance, and making that layer the place where agents come to consume data rather than replicating data outward. Partnering with Databricks, Workday built a universal data layer on Apache Iceberg™ governed end-to-end by Unity Catalog, bringing agents to the data rather than replicating data to agents.