To truly understand the unfathomable impact of AI on the enterprise, it’s essential to closely follow the latest research into how and why companies are using the technology. That’s why Cloud Wars has consistently shared research from companies that have heavily invested in the outcome and precision of AI tools.
For example:
- “The ROI of AI: Leaders Divulge Revenue Growth Levels, Top Use Cases Driving Wins“
- “Hybrid Workplace Future: AI Agents Gain Trust — But Not in Leadership Roles“
- “AWS Report Finds GenAI Overtaking Security Spending“
Now, Cloudera has released its latest findings on the impact of AI in a new report, “The Evolution of AI: The State of Enterprise AI and Data Architecture,” based on a survey of 1,574 IT leaders across the U.S., EMEA, and APAC regions.
Key Takeaways
The report focuses on the usage, adoption, and obstacles to AI implementation, as well as how they have evolved since the inaugural Cloudera survey a year ago. Now, 96% of respondents said that AI was “at least somewhat integrated into their core business processes.”
While a relatively small number — 21% — said AI was already fully-integrated into core business processes, 54% qualified current AI integration as significant. Regarding AI models, the most common were generative (60%), followed by deep learning (53%), and predictive (50%). The report concluded that as confidence in AI usage grows, so do the types of use cases and model architectures, such as agentic, which was already in use by 36% of respondents.
With data architecture playing such a critical role in AI enablement, the report also centered on this area. It found that, in terms of capability, 52% of those surveyed stated that integrated AI and ML operations tooling was their top requirement, followed by automated data pipeline orchestration (51%), granular data governance (44%), and a unified data access layer (41%).
AI Agent & Copilot Summit is an AI-first event to define opportunities, impact, and outcomes with Microsoft Copilot and agents. Building on its 2025 success, the 2026 event takes place March 17-19 in San Diego. Get more details.
What’s Holding Enterprises Back?
The report also examined the challenges faced by enterprise users. Regarding data architectures, users reported several technical limitations, with data integration being the most common issue, affecting 37% of respondents. Other frequently mentioned problems included storage performance (17%), compute power (17%), lack of automation (17%), and latency (12%).
Another significant factor highlighted by the report was cost. In fact, in last year’s survey, only 8% of respondents said the cost to access computation for model training was too high. Now, that has increased to 42%.
Beyond data and compute costs, the primary barrier to AI adoption, identified by 46% of respondents, has been security and compliance risks. Specifically, 50% of those surveyed cited the risk of data leakage during model training as a concern, followed by unauthorized access to data at 48% and insecure third-party AI tools at 43%.
Closing Thoughts
For me, one of the most significant points made in the report is this:
“Successful AI initiatives require businesses to assert control over 100% of their data, in all forms.”
Everyone knows that data architecture plays a crucial role in AI. However, the statement regarding the critical importance of having all of your data controlled and governed is significant. In many ways, exploring the obstacles to AI adoption is as important as examining the success stories. If there’s one key takeaway for users, it is the importance of organizing their data estate before embarking on a wide-ranging AI strategy.
Confidence in and understanding of AI is clearly growing at an exponential rate. Yet, to keep pace with the rapid development of AI, particularly in the agentic AI space, business leaders must first establish a solid data foundation. Additionally, for vendors, it is essential to provide robust data architecture that gives these business leaders the confidence to choose their products and thrive in the AI era.
Ask Cloud Wars AI Agent about this analysis
