UK AI Spending Rises on FOMO Despite Weak Results

Research commissioned by network provider Expereo and carried out by IDC found UK companies are increasing AI investment largely because of fear of being left behind. The study reported only 19% of AI projects exceeded expectations, and respondents pointed to poor training data and weak networks as key factors limiting returns.

The IDC survey polled organizations worldwide and found about 70% are investing heavily in AI tools and solutions. Half of respondents reported AI or machine learning will be a priority for effort or funding over the next 12 months. One in five admitted they are investing aggressively with little assessment of likely return on investment.

The impulse to invest out of competitive pressure varied by region. In the Asia-Pacific region, 37% of respondents cited fear of being left behind as a primary driver, compared with 13% in Europe and 10% in the United States. A separate analysis by software firm Orgvue found 57% of organizations in the US, UK, Canada and Australia had deployed AI mainly because competitors had, and 78% reported projects failed or remained stuck in pilot stages.

On outcomes, only 19% of survey respondents said AI projects exceeded expectations and just 5% said they significantly exceeded them. Slightly more than half of respondents attributed shortfalls to inadequate or poor-quality training data. Forty-seven percent pointed to higher-than-expected costs or a failure to achieve an adequate ROI, and 46% reported AI did not perform as expected.

Infrastructure was flagged as a constraint. Among organizations whose AI implementations failed to meet expectations, just over a quarter reported inadequate network or connectivity performance contributed to the problems. More broadly, 54% of respondents reported they need more flexible, scalable networks, and 51% reported they require greater resilience and reliability to maximize uptime.

Ben Elms, Expereo’s CEO, said, “Every enterprise we speak to is investing in AI, yet the data shows a clear gap opening up between AI ambition and AI outcomes. More often than not, that gap comes down to the network underneath.”

Respondents also flagged longer-term risks. More than half cited emerging security risks as a potential threat to their AI deployments. Just under four in 10 raised concerns about losing track of AI-related costs and ROI once the technology becomes embedded across the business.

Executives and IT leaders identified data quality, cost management, security and network readiness as obstacles to moving AI pilots into regular operations.