UK businesses are increasing their spending on artificial intelligence at an unprecedented rate, but escalating salaries for scarce talent and uncertainty over returns are prompting concerns about the sustainability of that investment.
A new survey suggests AI budgets will grow by 36 per cent this year, with average annual spending now exceeding $1 million. Yet the talent pool is tightening. Salaries for AI-related roles are expected to rise by 15 to 25 per cent in 2025, with many skilled professionals earning between $150,000 and $200,000. Thirty-five per cent of companies cite rising compensation as the single biggest barrier to filling AI positions.
Skills shortages remain a structural issue. More than a third of organisations say they cannot find suitably qualified candidates, particularly in cloud and data engineering, while 13 per cent admit they lack the in-house expertise to assess applicants effectively. This gap is slowing project delivery and undermining the ability of businesses to scale AI initiatives.
Ambition outpacing capability
Even as budgets climb, confidence in measuring value is lagging. Only half of respondents say they are confident in evaluating the return on AI projects, leaving a growing mismatch between investment levels and the ability to prove impact. Analysts warn that without clear metrics and governance, organisations risk committing significant sums to programmes that may not deliver measurable benefits.
Stuart Harvey, chief executive of data management firm Datactics, said the issue is not simply one of cost. “AI success doesn’t come from throwing money at tools alone, but by the people who build, manage, and govern those systems effectively,” he said. Harvey added that companies often prioritise specialist hires over the fundamentals of governance, explainability and data quality, increasing costs while reducing clarity around returns.
Building sustainable capability
Rather than entering salary bidding wars, the report suggests organisations should focus on long-term capability by upskilling existing teams in data quality, model oversight and governance. AI performance depends heavily on the accuracy and integrity of the data it processes, making in-house expertise in data preparation and structuring critical for scalability, compliance and trust.
Harvey argued that the businesses best positioned to succeed in the next phase of AI adoption will not necessarily be those with the largest budgets, but those with disciplined, data-driven foundations. “Without strong data foundations, even the most advanced models will fail to deliver meaningful results,” he said.
Industry analysts expect a growing emphasis on transparent, auditable AI systems that can withstand regulatory and commercial scrutiny. However, without investing in internal skills and robust data infrastructure, organisations risk inflating AI costs without improving outcomes.
As AI spending accelerates, the challenge for boardrooms will be to balance investment in tools with the less visible but equally essential investment in people and processes. The pressure to demonstrate value is likely to intensify, and the companies that can link their AI ambitions to measurable, sustainable results will be the ones that justify the rising cost of participation in the AI race.
