AI Agents Force Enterprises to Rebuild Data Foundations

Indian enterprises are encountering challenges in scaling AI from pilot projects to production. While 73% of surveyed businesses report measurable returns from AI investments, only 4% believe their enterprise data is fully ready for AI at scale. This readiness gap necessitates addressing years of accumulated data debt.

The data landscape is often fragmented, with systems built for disparate teams and use cases. AI agents, expected to reason across an organization, require not just access to data, but also an understanding of its meaning and context. This entails integrating scattered information, identifying real-time data needs, ensuring agents can comprehend context, and managing access permissions.

Inc42, in partnership with Oracle, hosted a roundtable discussion at the CTO Summit 2026 focused on building data foundations for AI. The discussion highlighted that business needs and intended outcomes should guide data architecture, agent access, and deployment costs.

Participants emphasized defining the problem to be solved before selecting technology. "Start from an outcome focus and then figure out what to build," stated Syed Mohd Bilal, Head of Data Science at Perfios. The unpredictable nature of AI agents necessitates more flexible data foundations that can support both known and future applications.

Furthermore, building context into data is critical for AI agents to understand its structure and meaning. A semantic layer, interpreting data's significance, is increasingly vital for agent reasoning. Simultaneously, governing data usage and establishing access controls are essential for ensuring security and compliance.