Mastering the data challenge in AI
Enterprises already have the data needed to improve their AI—but it’s fragmented, siloed and spread across disparate systems. In this report, we explore how AI leaders are forging a unified data view to build AI that’s not just functional but exceptional.

The advent of generative AI has unlocked a wave of spending by corporations across the world. In the US alone, tech companies plan to invest a heady US$1trn over the next four years to deliver the data centres, chips, energy and infrastructure to support AI, while businesses in every sector are investing in expert staff, computing power and AI models.[1]
If companies are to reap the benefits of this breakthrough technology, however, they first need to corral their various data sources into manageable systems so that AI can be put to work on them. Yet obstacles such as data silos and complex data engineering architectures often prevent organisations from achieving a unified view of their data.
In this study, we found:
Most organisations struggle to effectively connect AI to their full business data ecosystem. Just 29% say their existing architecture can fully do this.
Data silos remain a significant organisational barrier for AI. Respondents rank dealing with data silos and lack of integration as one of their two biggest challenges in managing AI data.
Businesses point to inadequate AI governance as a major risk to data integrity. Nearly 80% of respondents cited insufficient observability and monitoring as one of the top three contributors to poor data quality at their organisation.
To explore how businesses are tackling these challenges, Economist Enterprise, supported by Databricks and SAP, independently researched and wrote this report. The insights are based on cross-tabulated analysis of a subset of survey respondents from our Unlocking enterprise AI study published in November 2024.[2] The selected group comprises 450 executives and technologists across the financial services, healthcare and pharmaceuticals, and retail and consumer goods sectors.
Insights are also shared from several of our 2024 interviews with chief information officers and chief technology officers at leading organisations across these sectors, including Novartis, Flo Health, Mahindra Group, Thermo Fisher Scientific, and Providence.
References
- 1.Deloitte, “Can US infrastructure keep up with the AI economy?”, 2025, : https://www.deloitte.com/us/en/insights/industry/power-and-utilities/data-center-infrastructure-artificial-intelligence.html
- 2.Economist Enterprise’s 2024 Unlocking enterprise AI study combines a global survey of 715 technical executives and 385 data and artificial intelligence (AI) technologists who work across the fields of data engineering, data science and enterprise architecture. It also features insights from interviews with 28 C-suite executives from leading organisations across 11 industries. : https://insights.economistenterprise.com/new-globalisation/unlocking-enterprise-ai
