UK Public Sector AI ambitions hindered by fragmented IT environments
London, United Kingdom — 11th June 2026: New data from SolarWinds reveals that AI is being rolled out across the UK public sector on top of already fragmented systems. This is creating a challenging foundation for organisations looking to scale the impact of new AI tools.
In a survey of 1,040 global IT professionals*, more than half (57%) of public sector respondents operate in hybrid environments, with a further quarter (25%) primarily on premises. This creates a highly mixed foundation for new technologies to sit on. At the same time, nine in ten (90%) report fragmented IT tooling across systems and platforms, making it harder to deploy AI consistently and at scale.
This complexity is already being felt in day-to-day operations. More than a quarter (26%) cite integrating AI into existing systems as a key source of increased complexity, while 69% say they need to double-check AI outputs, and 67% struggle to trust its recommendations. A further 31% say keeping up with AI tools and capabilities is a major challenge, highlighting the difficulty of layering new technology onto systems that are still difficult to integrate and manage.
In parallel to this, organisations are under pressure to move quickly. Around one in six (16%) say they feel pressure to adopt AI faster than their organisation is ready for, while 42% say better infrastructure would help teams adapt to AI demands. Together, this points to a growing gap between AI ambition and the underlying systems needed to support it effectively.
Commenting on the findings, Rich Giblin, Head of Public Sector and Defence at SolarWinds said: "AI is creating real opportunities for public sector organisations to work more efficiently, improve services and make better use of their data. But only if the right foundations are in place first. The real risk is moving faster than your data, systems and processes can support.
"If AI is being introduced into environments where information is siloed, systems don't connect cleanly, or knowledge is outdated, then the technology will reflect those weaknesses rather than fix them. In practice, that can mean unreliable outputs, poor recommendations and much more effort spent checking and validating results.
"That's why the safest and most effective approach is to start with well-defined use cases in areas where organisations have stronger data, clearer processes and better oversight. From there, AI can be expanded more confidently. But if public sector organisations try to scale adoption before the groundwork is done, they risk limiting the value AI can deliver across services."
To learn more about the findings, visit: https://www.solarwinds.com/campaign/it-trends