How Hidden Cloud Storage Fees Became a Growing Enterprise Cost Risk

A new survey highlights how cloud storage has quietly evolved into a major financial pressure point for modern enterprises. As AI rapidly drives up data demand, the report finds hidden fees and operational challenges that are reshaping how organizations plan their future infrastructure.

According to Wasabi’s 2026 Cloud Storage Index report, organizations continue to face high and unpredictable cloud storage costs, with half of their total storage bills consumed by various fees (such as API calls, egress charges, and data operations) rather than the storage capacity itself. In 2025, nearly half of all organizations (49%) exceeded their cloud storage budgets, and this trend is largely driven by unexpectedly high data operations and API-related costs that contributed significantly to budget overruns.

“One of the results of this fee-heavy pricing structure is budget excess,” Wasabi explained. “84% of UK respondents cited at least one fee-related reason why their spending on public cloud storage exceeded budget expectations.”

AI investments are putting increasing pressure on data infrastructure

This report mentioned that these rising costs come at a critical time, as many organizations are increasing their investments in AI and transformation projects. A majority of organizations (59%) expect their AI infrastructure budgets to grow over the next year, with most of this investment directed toward data, storage, and compute resources rather than software.

However, only 32% of organizations report achieving positive ROI from their AI initiatives, though optimism is high, as 51% anticipate realizing positive returns within the next 12 months. This timeline is intensifying pressure on AI vendors and internal teams to accelerate performance, streamline workflows, and deliver meaningful outcomes quickly.

Data challenges are slowing down AI success

Organizations are increasingly deploying a diverse range of AI applications, with leading use cases spanning security monitoring, compliance analysis, anomaly detection, virtual agents, predictive analytics, and various recognition technologies. However, data-related issues remain the primary obstacles to effective AI implementation. 47% of organizations report challenges related to storage (such as cost, access, or data movement) while another 46% struggle with data quality and preparation efforts required to support AI workflows.

Most organizations believe that between 25% and 74% of their stored data qualifies as “dark.” Dark data is information that remains unindexed, unanalyzed, or otherwise inaccessible, including items like logs, recordings, legacy archives, and older documents. With AI initiatives increasingly dependent on clean, well‑governed data, an overwhelming 91% of respondents say that analyzing and operationalizing this dark data has become a strategic priority for their organizations.

How organizations can reduce storage costs and improve AI readiness?

Organizations should prioritize greater transparency, control, and optimization across their data ecosystems to reduce unpredictable costs and improve AI readiness. This means reassessing cloud storage architectures to minimize excessive fees, embracing multicloud or hybrid models for flexibility and cost efficiency, and investing in scalable infrastructure that supports growing AI workloads.

At the same time, companies must tackle data‑related barriers by improving data quality, strengthening storage management practices, and actively converting “dark data” into usable, governed assets that can fuel analytics and AI initiatives. Organizations can align storage strategy, cost management, and data governance to better manage risks, accelerate ROI, and ensure their infrastructure is prepared for the next wave of AI-driven innovation.