Rebuilding the data stack for AI

2026-04-27 · Source: MIT Technology Review · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Cloud Computing & IT Infrastructure · Depth: Intermediate, extended

Summary

Many enterprises face significant obstacles in AI adoption due to fragmented and ungoverned data infrastructure, despite the rapid advancements in consumer AI tools. Bavesh Patel of Databricks and Rajan Padmanabhan of Infosys emphasize that the effectiveness of enterprise AI is directly dependent on the quality and accessibility of an organization's data. To bridge this gap, data must be consolidated into open formats, precisely governed, and made accessible across functions, moving beyond siloed SaaS platforms. This foundational work enables businesses to achieve measurable outcomes, automate workflows, and launch new business lines. Infosys and Databricks collaborate to provide frameworks, such as a 3M measurement framework (adaptability, business value, responsible), to help clients prioritize AI initiatives and ensure they deliver tangible value, avoiding "terrible AI" and enabling a shift from systems of execution to systems of action.

Key takeaway

For CTOs and AI Architects aiming to scale AI within their organization, prioritizing a unified, governed, and open data architecture is paramount. Your AI's effectiveness directly correlates with data quality and accessibility, so invest in cleansing, organizing, and securing your data estate first. This foundational work, supported by a clear value roadmap, will enable repeatable, high-impact AI deployments and mitigate the risk of project failure.

Key insights

Effective enterprise AI hinges on unified, governed, and accessible data infrastructure, moving beyond fragmented legacy systems.

Principles

• AI quality depends on data quality.
• Data must be consolidated into open formats.
• AI initiatives require clear business value alignment.

Method

Implement a 3M measurement framework (adaptability, business value, responsible) to prioritize AI projects. Establish a unified data core with robust governance, leveraging tools like Unity Catalog for discovery, access, and semantics.

In practice

• Analyze your data estate for critical data consolidation.
• Develop an AI value roadmap tied to business metrics.
• Utilize open data platforms to avoid vendor lock-in.

Topics

• Data Stack Rebuilding
• Enterprise AI Readiness
• Data Governance
• Databricks Platform
• Agentic AI

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

Related on AIssential

• Modern Data Report 2026: What 500+ Data Leaders and Experts Say on AI-Readiness — Enterprise data foundations are insufficient for AI, hindering automation and decision-making due to issues with trust, context, and discoverability.
• Data quality is the AI strategy — Prioritizing data quality at the source is fundamental for effective AI and real-time decision support in healthcare.
• Stop building data products. Start building data services. — Traditional data product models hinder rapid growth and AI agent consumption; shift to open, governed data services.
• Understanding the data core: From legacy debt to enterprise acceleration — AI scalability and reliability depend on reinventing the data core, not just upgrading technology.
• Your AI Strategy Isn’t Failing Because of the Model — It’s the Data — Data governance and quality, not AI models, are the primary bottlenecks preventing enterprise AI from scaling to production.
• How AI Is Reshaping Enterprise Data Governance — AI is transforming data governance from reactive control to predictive, dynamic, and autonomous management.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MIT Technology Review.