Across industries, we’re seeing the same pattern play out: AI pilots look great on slides, impress leadership, and generate early wins… but fall apart the moment you try to scale them across the enterprise.

If this sounds familiar, you’re in good company. Gartner’s latest note on generative AI entering the “trough of disillusionment” simply validates what CDOs already know: the stall isn’t a technology problem. It’s a data and operating-model problem.

Most organizations are trying to build AI on foundations that were never designed for the speed, autonomy, and scale AI requires. The root cause is twofold: rushing ahead with weak data foundations and attempting to scale AI with a centralized model built for yesterday’s analytics.

This creates a mismatch between what enterprise AI needs and how the organization actually functions. And no amount of new tooling will paper over that gap.

To restart momentum and scale AI responsibly, CDOs must tackle three areas head-on: governance, operating model, and the evolution of their own role.

1. Beyond quality: Governance, lineage, and accountability are non-negotiable

Many organizations fixate on data quality alone. But quality — on its own — doesn’t make data AI-ready. AI requires trust, and trust is earned through governance, full lineage, and crystal-clear accountability.

The rise of explainable and regulated AI makes this unavoidable. When a regulator or an executive asks, “Why did the model make that decision?” you can’t hand-wave. You must show:

• Where every input came from
• How it has been transformed
• Who owns it
• Why the model behaved the way it did

Without enforceable lineage and governance baked into your pipelines, you’re not just risking AI failure; you’re risking flawed decisions, compliance breaches, customer mistrust, and operational chaos. Good data without good governance is still bad data for AI.

2. The centralized model: Yesterday’s answer, today’s bottleneck

Getting data AI-ready is proving far more complex than early pilots suggested. But the biggest blocker isn’t infrastructure, it’s the operating model.

The traditional centralized CDO model was built for reporting and BI. It cannot keep up with AI, where domains need to move at machine speed with their own quality standards, compliance requirements, and operational realities. A central team managing every schema change, access request, data interpretation, and quality rule becomes the choke point, and the more AI you deploy, the worse this gets.

This is why Actian’s research shows more than half of organizations shifting toward federated and center-led models. Policies alone aren’t enough; ownership must move to the domains.

Breaking the bottleneck requires the CDO to stop trying to control everything centrally and start enabling the system as a whole.

3. The CDO’s mandate: From gatekeeper to value multiplier

The decentralized model empowers CDOs to multiply their impact by placing accountability with the domain teams who understand their data. By embedding governance from the start — a ‘governance by design’ approach — organizations build trust and agility into every AI initiative.

Data contracts provide unified standards that establish trust and accountability across federated data environments. Each contract defines data structure, quality rules, access policies, and compliance requirements. When applied to AI agents, these standardized specifications enable AI agents to consume data safely while enabling compliance. In essence, data contracts become the new API for defining quality, usage, and trust.

When AI agents consume data autonomously at machine speed, proactive quality monitoring and prevention become imperative. That’s why data observability must shift left. Instead of reactively discovering quality issues downstream in data warehouses, identify and remediate them in operational systems at the source. 

Putting the principles into practice

Here’s how this transformation actually works. The CDO office maintains global governance standards applied universally across the organization, such as PII handling, HIPAA compliance, and data retention. Beyond these guardrails, each domain owns its data assets and defines domain-specific governance. For example, marketing might require strict consent tracking while finance needs data accurate to the penny with complete audit trails. 

The CDO evolves from central enforcer to chief architect, ensuring this federated system functions securely. This shifts focus from policing data access to empowering independent domain teams while maintaining unified, trusted standards and enterprise interoperability. 

This solid foundation enables what comes next: business process orchestration, where multiple AI agents communicate to accomplish complex workflows. Secure agent-to-agent communication is achieved through Model Context Protocol (MCP) servers using data contracts. With strong data foundations and decentralized operating models, multi-agent systems will deliver business value. Without them, multi-agents will remain theoretical.

Breaking through the AI stall

Scaling AI isn’t just about improving your data landscape. It requires rethinking how data is governed, owned, and operated across the enterprise.

Your ability to scale AI depends entirely on the foundation you build now. Organizations that get governance-by-design, federated ownership, and the new CDO mandate right will accelerate. Those that cling to centralized control and fragile foundations will stall.

The choice is clear, and the consequences will define your AI future.

About the Author:

As Chief Technology Officer at Actian, Emma McGrattan leads the company’s technology strategy, innovation, and product development in support of Actian’s mission to simplify how companies connect, manage, govern, and analyze data to transform businesses. Since joining the company three decades ago, McGrattan has played a pivotal role in the evolution and advancement of its analytics, data integration, and data management solutions, including the Actian Data Platform.

A prominent figure in the database industry, McGrattan is known for her expertise in data architecture, query optimization, and cloud transformation. Her leadership and contributions to these areas are widely recognized, making her a respected voice at technology events.

Passionate about creating a sustainable, inclusive future for technology, McGrattan is a celebrated advocate for women in tech and an active mentor dedicated to fostering inclusive cultures within the industry.