Enterprises report stronger results from AI adoption, while hybrid data architecture emerges as key enabler of secure, scalable AI at enterprise scale
Cloudera, the only company bringing AI to data anywhere, released findings from its latest global survey report, The Evolution of AI: The State of Enterprise AI and Data Architecture. Polling more than 1,500 IT leaders, the report examines how AI adoption has accelerated, enterprise data architectures have evolved, and new challenges have emerged around securely scaling AI in 2025. Building on Cloudera’s 2024 study, the report highlights how priorities, obstacles, and goals have shifted in just one year, offering a snapshot of the rapidly changing enterprise AI landscape. Based on this report, it’s apparent that the future will be driven by AI, and AI is fueled by all data anywhere the data resides. Organizations need access to 100% of their data, no matter where it is or what type, to securely govern it and derive robust real-time and predictive insights without compromise.
The most striking finding: 96% of IT leaders reported AI is at least somewhat integrated into their core business processes. This is up from 2024, when 88% of survey respondents said they were currently using AI within their companies. This indicates that AI has gone from experimentation to full integration in core processes and workflows. And it’s paying off: 70% of respondents said they have achieved significant success with AI initiatives, with only 1% having yet to see results.
Also Read: CIO Influence Interview with Dipto Chakravarty, Chief Product and Technology Officer at Black Duck
Enterprises are tapping into a variety of AI forms to achieve these results, including generative (60%), deep learning (53%), and predictive (50%). The confidence to diversify AI portfolios is also up, with 67% of IT leaders feeling more prepared to manage new forms of AI, particularly AI agents, than a year ago. Behind this success is a growing shift in how enterprises approach data. A hybrid approach to data architecture has become the norm, offering organizations the flexibility to manage AI across cloud and on-prem environments. When asked about the biggest advantages of a hybrid approach, respondents cited security (62%), improved data management (55%), and improved data analytics (54%).
Despite progress, enterprises acknowledge they are still on the journey toward maximizing AI’s potential and ROI. While 24% said their culture is now extremely data-driven, up from 17% last year, most recognize more work is needed to embed data-first thinking into business practices. The biggest technical limitations identified in current data architectures when supporting AI workloads include data integration (37%), storage performance (17%), and compute power (17%). Data accessibility is another hurdle: only 9% of organizations said all of their data is available and usable for AI initiatives, while 38% reported that most of their data is accessible.
“In just a year, AI has shifted from a strategic priority to an urgent mandate, actively reshaping operations and redefining the rules of competition,” said Sergio Gago, Chief Technology Officer at Cloudera. “But our survey shows that enterprises still face deep challenges around security, compliance, and data utilization, with many getting stuck at the proof-of-concept stage. Cloudera’s mission is to bring AI to data wherever it resides—public cloud, private cloud, and on-premises—while ensuring full governance, lineage, and trust. With innovations like Private AI and secure, GPU-accelerated generative AI behind the firewall, we give enterprises the full control and confidence to unlock insights from 100% of their data and accelerate adoption in this new Era of Convergence.”
Other key findings include:
- Enterprises are leveraging the cloud: When asked where their company’s data was stored, 63% of respondents said private cloud, 52% said public cloud, and 42% said data warehouse.
- Integrating AI is not without security concerns: Exactly half of respondents said data leakage during model training was a concern related to AI security, with 48% saying unauthorized data access, and 43% saying unsecure third-party AI tools.
- Despite those concerns, organizations feel confident: Almost a quarter (24%) of respondents said they were extremely confident in their organization’s ability to secure data used in AI systems, 53% were very confident, and 19% were somewhat confident.
Catch more CIO Insights: CIOs and the Inferencing Economy: Planning for a Future Where AI is Always On
[To share your insights with us, please write to psen@itechseries.com ]
