As enterprises move AI from experimentation to operations, the data foundation supporting these systems is emerging as a critical challenge. The latest Modern Data Report: 5 Emerging Trends in Enterprise Data & AI from The Modern Data Company highlights a growing gap between AI adoption and confidence in the data behind it.
The report draws on more than 540 responses from data leaders and practitioners across 65+ countries. It finds that 57.3% of organisations are already piloting AI agents or using them in production, while only 8.4% say the data feeding their AI systems is trustworthy enough for production. This highlights the gap between the pace of AI adoption and the readiness of the data foundations supporting these systems.
Data quality and trust have also emerged as the number-one barrier to putting AI agents into production, with 76% of respondents ranking them among their top three barriers. The findings are particularly relevant for Indian enterprises as AI moves from experimentation towards operational use across business functions. The growing use of AI agents across enterprise data, systems and workflows is raising the bar for trusted data, business context, governance and clear ownership.
According to Saurabh Gupta, President & CEO, The Modern Data Company, the report shows that the challenge is shifting from AI experimentation to operating AI reliably at scale. He noted that data quality and trust now rank among the biggest barriers to putting AI agents into production, while governance must extend to how AI systems use enterprise data and act on it. As AI takes on a more active role in business workflows, organisations need to establish the context, controls and accountability required for these systems to operate with confidence.
The report also highlights the growing importance of business context for AI. 61% of respondents say a reliable context layer is critical or very important for AI agents, yet only 16% say their organisation deliberately engineers its context layer as a product.
Business definitions, relationships, policies and other forms of organisational knowledge often remain distributed across systems and teams. As AI operates across the enterprise, this context increasingly needs to become an integral part of the data foundation.
The research also points to a shift towards more consolidated data environments. 47% of organisations are actively consolidating towards fewer platforms, while another 17% are evaluating consolidation. At the same time, 46% say more than a quarter of their team’s time is spent maintaining or integrating tools, highlighting the operational cost of increasingly fragmented data environments.
Governance is also expanding as AI becomes more operational. 62% rank security and governance concerns among their top three barriers to putting AI agents into production, while only 18% have a clear, documented AI accountability framework.
The report finds that organisations already running AI agents in production are 3.6 times as likely to have deliberately engineered their context layer. They are also three times as likely to report confidence in the data behind their AI compared with organisations interested in AI agents but yet to begin. The report notes that these relationships do not establish causation, although the pattern remains consistent.
For Indian businesses, Sanjoy Roy, Vice President, APAC, Middle East & EU, The Modern Data Company, said the next phase of AI adoption will focus on turning AI capabilities into measurable business outcomes. He emphasised that achieving this will require organisations to build trusted data and business context into their AI foundations from the outset.
The report is based on the third Modern Data Survey, an ongoing research initiative conducted through the Modern Data 101 community. The community brings together more than 15,000 members globally, while the current survey includes responses from data leaders and practitioners across 65+ countries.