D&B's database of 642 million businesses was built for humans, not AI agents. So they rebuilt it.

2026-05-22 · Source: VentureBeat · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Cloud Computing & IT Infrastructure · Depth: Intermediate, short

Summary

Dun & Bradstreet (D&B) rebuilt its Commercial Graph, a database covering 642 million businesses, to accommodate AI agents. Originally designed for human analysts, the system's fragmented architecture, static relationship tracking, and inability to handle sub-second latency queries became problematic as customers integrated agents into workflows. The database, which expanded to 642 million records with 11,000 fields each and processes 100 billion data quality checks monthly, required a fundamental shift. D&B's solution involved consolidating its databases onto cloud infrastructure, redesigning the schema, and implementing a data fabric to create a unified knowledge graph that tracks dynamic relationships. They also developed a structured access layer (MCP) for agents and a "Know Your Agent" identity verification system to authenticate agents and ensure entity consistency across complex multi-agent workflows.

Key takeaway

For Directors of AI/ML planning agent deployments, your existing data infrastructure likely needs a fundamental overhaul. You must prioritize consolidating fragmented data, designing for dynamic entity relationships, and embedding robust data lineage from the outset. Implement "Know Your Agent" protocols and integrate entity consistency checks into multi-agent workflows to prevent errors and ensure reliable, verifiable agent operations. Failing to address these foundational data challenges will severely limit your AI agent capabilities and scalability.

Key insights

Enterprise data systems built for humans require fundamental redesign for AI agents.

Principles

• Data foundations must be clean, normalized, and consolidated for AI.
• Design data systems for dynamic, not static, entity relationships.
• Embed data lineage from the start for agent-produced answers.

Method

D&B consolidated fragmented databases to cloud, redesigned schema, built a data fabric for normalization, created a unified knowledge graph, and added a structured access layer with entity resolution and agent identity verification.

In practice

• Implement "Know Your Agent" for machine authentication and entitlements.
• Integrate entity verification agents into multi-agent workflows.
• Use a business verification agent as a persistent reference point.

Topics

• AI Agents
• Enterprise Data Management
• Knowledge Graphs
• Data Modernization
• Data Governance
• Entity Resolution

Best for: CTO, VP of Engineering/Data, AI Product Manager, AI Architect, Data Engineer, Director of AI/ML

Related on AIssential

• Agents are ready but your architecture probably isn't — AI value stems from outcome-driven design, robust data infrastructure, and proactive governance, not just model selection.
• AI Agents and the Fight for Customer Data — Centralized, open data infrastructure is crucial for AI agents, despite vendor data access restrictions.
• The Data Agent Stack - Part 2: The Data Foundation Is the Agent — The reliability of data agents hinges on a well-governed data foundation that explicitly encodes meaning, quality, and ownership for machine consumption.
• The TechBeat: The IDE Is No Longer Just a Place to Write Code (6/17/2026) — AI is deeply integrating into development, operations, and business strategies, demanding new approaches.
• Three insights you may have missed from theCUBE’s coverage of Snowflake Summit 2026 — Enterprise AI's next wave demands robust data foundations, new security frameworks, and governed intelligence for successful production deployment.
• The Production AI Playbook: Deploying Agents at Enterprise Scale — Sandipan Bhaumik, Databricks — A structured five-pillar framework is essential for successfully deploying and managing enterprise-scale AI agents in production.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by VentureBeat.