AI has advanced rapidly, yet only a small group of enterprises are turning early wins into meaningful profits. Most have proven that generic AI can increase productivity and accelerate workflows, but very few have laid the foundation needed to scale that impact across an entire business. The moment facing senior technology leaders is decisive. The differentiator is no longer making progress alone, but rather whether the data, governance and architecture are mature enough to translate AI momentum into enterprise-wide performance.
How are enterprise leaders thinking about this change? We’ve partnered with MIT Technology Review Insights to uncover the biggest trends and changes in enterprise AI strategies. Read insights from 800 senior data and technology executives on what it takes to build a high-performing data and AI organization.
Data and governance drive high-quality AI
The 2025 MIT Technology Review highlights the many organizations creating research infrastructure. One organization that found success in AI through this approach is Fox Corporation, which created Sports AI, a multi-modal chatbot capable of answering sports questions using live commentary and journalistic content. However, the team discovered that their legacy search foundation could not support the level of accuracy required. This constraint led them to rebuild the backend using a semantic search architecture that could interpret content contextually and deliver it to the right model. This investment in data context, lineage, and model orchestration produced measurable improvements in performance and user experience.
This story is a reminder that competitive differentiation increasingly comes from the data and governance layers beneath AI, not just the models.
At Databricks, we see this pattern across many of the global enterprises we work with. Organizations making real progress are investing in unified data governance, semantic context, and a simplified architecture that allows models and agents to operate on trusted data.
Differentiator: Integrated Data, Analytics and AI
In MIT research, a trend is clear. Enterprises that integrate data, analytics, and AI on an integrated basis gain the ability to scale with speed, reliability, and confidence. Those that remain fragmented still experience friction: inconsistent control, unclear lineage, and disjointed governance patterns.
None of these challenges are insurmountable. In fact, many organizations already have the ingredients for success. They have capable analytics teams, modern cloud environments, and mature data platforms. What is changing now is executive intent. Leaders are prioritizing coherence, clarity, and cross-functional alignment as gateways to enterprise-wide AI performance.
Across our client base, the same signal is consistent. When teams integrate data, analytics, and AI on a unified basis, they remove friction and gain the reliability needed to scale.
Preparing for change in agentic AI
This foundation-first mindset becomes even more important as organizations explore agentic AI. While generative AI focuses on generating content or insights, agentic AI relies on goals, context, and the ability to take informed action. This makes governance, lineage and risk management essential rather than optional.
Enterprises that have begun this transformation are considering agentic capabilities as a catalyst for discipline. Workday, for example, focuses heavily on presenting the right data to agents, validating the authority behind agent actions, and ensuring that governance patterns are consistent at every level. Their approach reinforces that responsible autonomy can only be achieved when data strategy and AI strategy go hand in hand.
3M offers another perspective. Their data and AI teams focus on building deep metadata and business context before expanding agentic capabilities. By strengthening the semantic layer behind their data, they ensure that every model and agent has the clarity they need to make trusted decisions. For them, context is not a technical detail, but a strategic asset.
Turning a Data Foundation into Profit
The organizations growing the fastest aren’t waiting for perfect conditions. From our work with CIOs, CTOs, and CDOs, the organizations that move fastest are those that simplify architecture, centralize governance, and treat data context as a strategic asset rather than a technical convenience. Their progress shows that responsible scaling is not a barrier. This is what unlocks what allows AI to perform reliably in production and differentiate the leaders from the rest of the field.
As executives plan for the next decade of AI innovation, the real question is no longer whether AI will transform their business. It is whether their organization’s data, governance and architectural foundations are ready to support autonomy, action and long-term performance.
Where to go deeper
Download the full MIT Technology Review for a detailed look at the practices that differentiate high-performance data and AI organizations from their peers.
Watch the on-demand webinar: Unlocking the Future of Data and AI to learn how leaders from 3M, Workday, Reckitt and Databricks are aligning data, governance and AI to drive real results.
