95 Percent Of Organizations Delayed AI Projects. Governance Was Among The Biggest Roadblocks

A new Cloudera survey suggests that governance, compliance and data architecture challenges may be creating roadblocks to AI success than access to the technology itself.

Most organizations no longer struggle to access AI tools. Instead, many struggle with deploying them as they face several mounting roadblocks.

A new Cloudera survey of 1,500 enterprise architects, cloud infrastructure leaders, and data architects found that 95 percent of organizations have delayed or outright canceled AI projects this past year due to issues with data governance, compliance or regulations.

The findings suggest that AI adoption may be moving faster than the rate at which organizations can effectively govern and manage the data that powers it.

The report, released Aug. 11, describes what Cloudera calls "The Great AI Re-Architecture," a period in which organizations are rethinking data architecture, governance frameworks and infrastructure to support AI at scale.

According to the report, 77 percent of organizations are already actively using AI, yet 72 percent said their current data architecture "requires a significant overhaul to meet future AI requirements."

While legacy infrastructure plays a big role, the survey suggest governance poses as significant a challenge.

Nearly three-quarters of respondents said AI has made data governance more complex, while 55 percent reported delaying or canceling more than six AI projects over the past 12 months because of governance, compliance, or regulatory issues.

For midmarket IT leaders, the findings reinforce a reality that often becomes apparent once organizations move beyond the AI pilot phase.

Deploying a chatbot, copilot or AI-powered app is only part of the equation. Organizations must also determine who can access data, how information is secured, how compliance is met, and whether governance policies can keep pace with rapidly evolving AI capabilities.

Meanwhile, organizations are increasingly reconsidering where AI workloads run.

According to the survey, 66 percent moved AI workloads from public cloud environments back to private cloud or on-premises infrastructure during the past year. Another 25 percent said they plan to prioritize a hybrid-first architecture strategy over the next two years.

That very shift may reflect a growing realization that AI success depends less on adopting a specific model, and more on building data environments that balance security, governance, performance and costs.

For organizations with stalled AI initiatives, Cloudera's report suggests the bottleneck isn't a lack of AI tools or support from executive leadership. Instead, the challenges are with governance frameworks, compliance control and the data architecture required to deploy AI at scale.