Artificial intelligence can make weaknesses in an organisation’s data harder to ignore. As teams move from experimentation toward wider adoption, problems with quality, structure, accessibility and ownership can quickly affect the usefulness of AI systems. A model may be sophisticated, but its outputs still depend on the information available to it and the controls around how that information is managed.
This is why data foundations are becoming a strategic issue rather than a technical housekeeping task. Organisations that want to use AI across more processes need to understand what data they hold, where it sits, how it moves and whether people can trust it. Without that clarity, AI programmes can become difficult to scale, difficult to govern and difficult to connect to measurable business outcomes.
AI raises the cost of unclear data
Many organisations have accumulated data across different systems, teams and platforms over time. That is manageable when information is used for a narrow set of reports or operational tasks. It becomes more complicated when AI systems start drawing on multiple sources at once, especially when those sources use different definitions, formats or access rules.
In that environment, quantity is not the same as readiness. More data can create more noise if the organisation cannot explain which sources are authoritative, how records are maintained or who is responsible for resolving inconsistencies. AI can expose these issues faster because it increases the number of decisions, recommendations and automated actions that may depend on the same underlying information.
Good foundations connect data work to business outcomes
Data programmes can lose focus when they are treated as large technology projects without a clear connection to the outcomes the organisation is trying to achieve. AI makes that risk more visible because the pressure to experiment can encourage teams to launch use cases before defining what success should look like.
A stronger approach begins with strategic intent. Leaders can identify two or three priorities that matter over the next 12 to 24 months, such as improving customer insight, increasing the visibility of operational performance or making better use of existing assets. Those priorities can then guide decisions about which data needs attention first.
Infrastructure choices shape what AI can do
Data quality is only one part of the foundation. Infrastructure decisions also affect how quickly information can be accessed, combined and used. Choices about cloud, on-premise and hybrid environments influence scalability, cost, security and the speed at which teams can work with data.
The right architecture depends on the organisation’s goals and constraints. A highly regulated business may have different requirements from a digital-first company. A company with large volumes of operational data may face different challenges from one whose most important information sits in customer or financial systems.
Governance needs to evolve with adoption
As AI use expands, governance cannot remain limited to a central technology team. Business, operations and technology functions may all make decisions that affect how data is collected, interpreted and used. Without shared standards, different teams can create conflicting definitions, duplicate work or introduce new risks.
Cross-functional governance provides a way to make those decisions more deliberately. It can clarify who owns important datasets, how quality issues are handled, what access is appropriate and how changes are approved. It can also help organisations treat data maturity as an ongoing capability rather than a one-off programme.
Strong data foundations make AI easier to scale responsibly
AI adoption is likely to place greater pressure on organisations to make data more usable, trusted and accessible. A structured approach to AI and data governance can help frame the practical questions around quality, access, ownership, infrastructure and the outcomes that data is expected to support.
The goal is not to make every dataset perfect before using AI. That can delay useful progress and create a programme that is too broad to manage. The more practical objective is to create enough clarity and control around the data that matters most, then improve the foundation as adoption expands.
Organisations that take this approach can make AI decisions from a stronger starting point. They can connect use cases to clearer business priorities, identify the data constraints that matter most and make governance part of normal decision-making. As AI becomes more deeply embedded in operations, those foundations are likely to matter as much as the models and tools built on top of them.
Copyright © 2026 California Business Journal. All Rights Reserved.
For California Business Journal Disclaimers, go to /terms-conditions/.
