Public sector AI ambitions outrun its infrastructure

TL;DR:

  • New analysis of Nutanix’s Enterprise Cloud Index finds 73% of public sector technology leaders say their infrastructure is not ready to run complex AI workloads on premises.
  • 91% warn that unvetted “shadow AI” use creates severe mission and security risks, as tools spread beyond approved governance.
  • The survey of 1,600 executives across 15 countries, including the UK, points to a widening gap between AI ambition and delivery capacity.

Public bodies are pushing AI from pilots into operational services faster than their systems and governance can safely support, according to industry analysis of Nutanix’s global cloud survey. The findings describe a familiar tension: appetite for AI outpacing the data, compute and controls needed to run it responsibly.

The readiness gap

The numbers point in one direction. Nearly three-quarters of public sector leaders say their infrastructure cannot yet handle demanding AI workloads on premises, and 87% expect their use of application containers to rise over the next three years as they modernise. The most acute worry is shadow AI — staff reaching for tools outside sanctioned channels — which 91% link to severe security and mission risk. Organisational silos between IT and service teams make the gaps harder to close.

The UK parallel

For British readers the pattern is recognisable. The survey spans the UK, and its picture of thin readiness sits alongside domestic evidence of shallow adoption: ONS figures this week found UK organisations using an average of just 1.6 AI technologies each. Whitehall’s ambitions are running well ahead of that baseline, and the machinery to deliver them is still being reshaped — the government has folded DSIT into a new structure with a dedicated AI minister. Live deployments such as the NHS App’s AI triage pilot show the upside when the foundations hold; the survey is a reminder of what happens when they do not.

Looking forward

The report frames regulated sectors — government, healthcare, financial services — as under greater pressure than organisations overall, because data sovereignty and compliance raise the stakes. The test for UK public bodies is whether governance and infrastructure investment catches up before shadow AI use hardens into an unmanaged norm across benefits, fraud detection and frontline services.

Share this article