More than half of UK enterprises are struggling to accurately measure the environmental impact of artificial intelligence, raising concerns about how organisations validate AI’s role in meeting sustainability commitments.
New research from Telehouse Europe found that 56% of UK IT decision-makers say their organisations cannot reliably track the emissions generated by AI workloads. The lack of accurate measurement is emerging as a major obstacle for businesses seeking to demonstrate progress against net-zero and wider environmental, social, and governance targets.
The findings point to a growing gap between ambition and evidence. Nearly nine in ten respondents, or 89%, believe AI is accelerating their organisation’s ability to achieve net-zero emissions and broader ESG goals. Yet without verifiable emissions data, those claims remain difficult to substantiate, leaving sustainability strategies exposed to scrutiny from regulators, investors, and customers.
The research also highlights a misalignment between long-term intent and day-to-day decision-making. While 79% of IT leaders say AI will be essential to meeting their organisation’s 2030 sustainability targets, fewer than half, 45%, currently rank sustainability as a top priority when allocating AI-related spending. Productivity considerations dominate investment decisions, cited by 62% of respondents, followed by cost at 49%.
That imbalance is further reflected in infrastructure choices. Only 23% of IT decision-makers said renewable energy mix and energy efficiency were key factors when selecting an AI infrastructure partner, despite the growing energy intensity associated with AI training and inference workloads.
Alongside sustainability challenges, the research raises questions about operational resilience as AI becomes more embedded in core business processes. While 93% of respondents said geopolitical uncertainty is influencing infrastructure strategy, preparedness appears uneven. Nearly six in ten, or 59%, reported that it would take more than an hour to switch AI inference workloads to an alternative site if their primary UK AI data centre were taken offline by a regional incident.
Mark Pestridge, Executive Vice President and General Manager at Telehouse Europe, said confidence in AI’s sustainability benefits is outpacing organisations’ ability to prove them. “AI has earned its place in sustainability strategies but belief in its potential is running ahead of evidence. Firms are confident in AI’s contribution to efficiency, yet far less equipped to measure or verify its environmental cost,” he said.
Pestridge added that improving carbon visibility will be critical if enterprises are to move from aspiration to accountability. “Until sustainability becomes a core criterion in spending and infrastructure decisions, green ambitions will remain more aspiration than proof.”
The findings come as UK enterprises face increasing pressure to demonstrate credible progress on sustainability, particularly as AI-driven workloads place growing demands on energy-intensive infrastructure. For many organisations, the next phase of AI adoption may hinge less on performance gains, and more on the ability to measure, manage, and transparently report its environmental footprint.
You must be logged in to post a comment.