AI Overcapacity Concerns Rise as 79% of Companies Use GPUs at Half Capacity or Less

Concerns about overcapacity in artificial intelligence infrastructure are growing, with 79% of surveyed companies reportedly using their graphics processing units (GPUs) at half capacity or less. Despite the low utilization levels, 70% of companies still plan to adopt new AI computing platforms, while a lack of finance-reviewed return-on-investment estimates raises questions about spending decisions.

The figures were highlighted by Coin Bureau, citing VentureBeat, in a report examining the gap between current AI computing capacity and companies' plans for further investment. The data points to a potential mismatch between existing infrastructure utilization and continued demand for additional computing platforms.

Low GPU Utilization Raises Questions About AI Spending

The finding that 79% of surveyed companies use their GPUs at half capacity or less suggests that many organizations may not be fully utilizing their existing computing resources. GPU capacity is an important consideration for AI workloads, making utilization levels relevant when companies assess whether they need additional infrastructure.

However, low utilization does not by itself establish that every company has purchased too much computing capacity. The reported figures do not specify why GPUs are underused, how utilization was measured, or whether the available capacity is intended for future workloads.

Companies Continue to Plan New AI Platform Adoption

Despite the utilization concerns, 70% of companies reportedly still intend to adopt new AI computing platforms. The figure highlights the continued appetite for additional AI infrastructure even as questions emerge about how effectively existing resources are being used.

A further concern involves financial oversight. According to the reported figures, 91% of companies planning to adopt new platforms lack return-on-investment estimates that have been reviewed by their finance teams. Without such assessments, it may be more difficult for organizations to evaluate the expected financial benefits of additional computing investments against their costs.

The figures do not establish that the planned investments will fail to generate returns, nor do they quantify the amount of spending at risk. They instead highlight two issues for organizations considering further AI infrastructure commitments: whether current GPU capacity is being used effectively and whether proposed investments have undergone financial review.

The central figures remain stark: 79% of surveyed companies use GPUs at half capacity or less, 70% plan to adopt new AI computing platforms, and 91% of companies planning new platforms lack finance-reviewed ROI estimates.

Writer: Victoria Hale
Technology & Blockchain Writer

Victoria Hale writes about blockchain technology, digital infrastructure, and the intersection of emerging technologies with finance. Her articles explore how new protocols and systems are shaping the evolving digital economy.

She prioritises clarity and accuracy when explaining technical developments to a general audience.