The Nutanix State of Enterprise AI Report, a global research study of 650 IT, DevOps, and Platform Engineering decision makers, found the vast majority of organisations (90%) placed a high emphasis on prioritising AI.

Despite the rapid embrace of AI as a competitive differentiator, almost the same percentage of respondents (91%) agreed their organisation’s IT infrastructure needs to be improved to more easily support and scale AI workloads.

Other key findings from the report include:

• Only 10% of organisations plan to build their own AI models, with the remaining 90% leverage existing, pre-trained AI models for their AI applications. A key driver to this finding may be due to the prevailing skills gap many enterprises are experiencing when recruiting AI talent.
• Security, reliability, and ease-of-management are the three key considerations informing infrastructure upgrades ahead of planned AI deployments. More than half (53%) of respondents mentioned data security as a key driver of AI application and infrastructure upgrades, while 52% listed infrastructure resilience and uptime, and 51% noting infrastructure management at scale.
• The need to improve the transfer of data between cloud, data centre, and edge environments to support AI data initiatives was a priority for more than half of respondents. With a growing awareness that data generated at the edge will be fundamental to any AI deployment, 83% of respondents indicated they plan to increase investment in edge strategies to support their AI initiatives.

Nutanix senior vice president product and solutions marketing Lee Caswell said to reap the full rewards AI has to offer, it was becoming increasingly clear these workloads needed the flexibility to run wherever they’re needed across the enterprise.

“We’re in a new era where success is defined by maximising AI’s potential,” Caswell said. “However, most of today’s infrastructures weren’t designed to handle the unique management and security needs of AI apps, especially as you move your AI workloads across different environments. Our view is clear: To accelerate your AI initiatives, you must maintain control of your data, privacy, and models.

“Hybrid multicloud platforms are fundamental to AI success, as they can allow you to run AI anywhere your business needs it, from the core right through to the edge.”