Wavestone surveyed 200 senior IT leaders across the UK's largest organisations and found 79% reporting strong or very strong business demand for AI. The same study found that 70% have not fully implemented the operating models needed to deliver on it. The gap is not between AI believers and sceptics — it's between ambition and execution infrastructure.
Most organisations have established formal foundations: 78% have human oversight frameworks in place, 78% have AI training programmes, and 77% have governance structures. Yet these remain inconsistently embedded in how technology services are actually designed and delivered. Governance, in particular, tends to arrive as a control layer applied after deployment rather than built into AI-enabled processes from the start — even though 94% of respondents expect governance investment to increase over the next 18 months.
Talent is the crunch point. Lack of AI-ready skills ranks as the primary challenge in the study, yet upskilling is the area most frequently under-prioritised. Third-party vendors are filling part of the gap: 87% of IT leaders say external vendors enable faster AI deployment, and 46% have moved to hybrid delivery models. The risk that comes with that is the slow accumulation of vendor dependency and the erosion of internal capability — knowledge that builds in the vendor's team rather than the client's.
Gonzalo Gonzalez, Associate Partner at Wavestone, said: "Many organisations treat AI as a technology upgrade, but don't execute the operating model transformation required to drive sustained value. Pilots often fail not because the AI isn't capable or technically brilliant, but because the right training, governance and change management aren't in place from day one. Organisations that succeed put the user first and build the right governance structure from the outset."
Bryn Barlow, Chief OpEx Officer at AXA UK, added: "AI-enabled transformation is one of the defining leadership challenges of our time. However, beyond technology, what determines whether organisations succeed is the human dimension. Psychological safety matters. Happy, confident, and inquisitive teams will ultimately have a far greater impact on successful AI adoption than technology, data, or observability alone."
The 70% execution gap is a useful data point for boards currently receiving optimistic AI forecasts from management. The operational foundations — skills pipeline, governance timing, the model for building internal capability alongside external vendor relationships — are what separates organisations that run pilots from those that run production at scale.
To stay across the latest in cloud, AI and enterprise tech analysis from Compare the Cloud, subscribe to our weekly newsletter at https://www.comparethecloud.net/newsletter
