Trust Your Data to Scale Agentic AI: Top CDO Investments
▼ Summary
– AI adoption in large companies has surged to 69%, up from 48% in 2025, with nearly half already using agentic AI.
– A major barrier to scaling AI is poor data quality and access, cited by 50% of those planning agentic AI adoption.
– Most data leaders report that governance and visibility have failed to keep pace with the rapid increase in employee AI use.
– A vast majority (86%) of chief data officers plan to increase investment in data management to support and secure AI initiatives.
– There is a critical workforce skills gap, as 75% of data leaders say employees need upskilling in data and AI literacy to use AI responsibly.
A new survey of chief data officers reveals a critical juncture for businesses scaling artificial intelligence. While adoption rates for generative and agentic AI are climbing rapidly, success hinges on the foundational quality, governance, and accessibility of data. The research indicates that nearly 70% of large enterprises are now using generative AI, a significant jump from previous years. However, this expansion is encountering substantial roadblocks directly tied to data infrastructure and employee competency.
The report highlights a pressing disconnect: although 65% of data leaders believe their employees trust the data used for AI, a staggering 75% of those same leaders say their workforce needs upskilling in data literacy. This paradox suggests that confidence in data may sometimes stem from a lack of understanding rather than from proven reliability. Without proper literacy, employees may struggle to identify poor-quality or untrustworthy information, potentially leading to flawed AI outputs and business decisions.
Governance frameworks are failing to keep pace with technological adoption, creating another major vulnerability. An overwhelming 76% of data leaders admit that their company’s visibility and control have not evolved in step with employee AI use. This governance gap poses significant risks for security, compliance, and ethical AI deployment. As AI systems, particularly autonomous agents, handle more critical tasks, the absence of robust oversight could result in unintended consequences and regulatory challenges.
Investment in data management is poised for a sharp increase, with 86% of CDOs planning to boost spending in this area. Their priorities are clear: enhancing data privacy and security, improving governance protocols, and fostering greater data and AI literacy across the organization. These investments are not merely operational upgrades but essential steps to unlock AI’s full potential and mitigate the risks associated with rapid deployment.
When it comes to the specialized field of agentic AI, where systems can autonomously execute tasks, adoption is already at 47%, with another third of companies planning implementation soon. The primary hurdles here are distinctly data-centric. Half of the organizations cite data quality and retrieval as the top barrier, followed closely by security concerns and a shortage of in-house expertise. Overcoming these challenges is paramount, as the promised benefits are substantial, ranging from enhanced customer experiences to improved analytical decision-making and streamlined regulatory compliance.
Scaling AI from limited pilots to full production integration fundamentally depends on confidence in data. Data reliability is viewed by 57% of leaders as the key obstacle to broader implementation. To build this reliability, organizations are focusing on three core actions: refining data and AI workflows, increasing investments in data quality initiatives, and strengthening metadata collection and management practices. These steps are designed to create a more trustworthy and actionable data foundation.
The path forward requires a balanced, multi-faceted strategy. Data leaders anticipate working with an average of seven different vendor partners for data management and eight for AI management, acknowledging that this ecosystem, while necessary, adds complexity. The overarching goal is to cultivate an environment where high-quality data and strong governance enable trustworthy AI applications. This involves not only technological investment but also a committed focus on human capital, ensuring employees possess the literacy to use these powerful tools responsibly and effectively. The businesses that succeed will be those that treat data not as a byproduct, but as the core strategic asset driving intelligent automation.
(Source: ZDNET)
