Education bodies boost AI budgets despite poor returns

Nearly half of education organisations plan to increase AI budgets this year, according to research from cloud storage company Wasabi. Yet only 37% of existing AI projects in the sector are delivering a positive return on investment.

The findings highlight a gap between spending plans and financial results, with infrastructure costs emerging as a major obstacle. In Wasabi's survey of 241 education respondents, 67% of AI infrastructure budgets were allocated to data, storage and compute.

Educational institutions are under pressure to expand AI use while managing the cost of the systems needed to support it. Half of the respondents identified data storage issues, including the cost of storing and accessing data, as the main challenge in implementing AI projects and services.

The study suggests that cloud storage charges are taking an increasing share of education technology budgets. Fee-related costs such as data egress, API operations and data access accounted for 54% of public cloud storage spending among education respondents, up from 50% a year earlier.

That shift has made storage spending less predictable for many institutions. Some 41% of respondents said they exceeded their public cloud storage budgets over the past year.

Cost pressures

The report comes as schools, colleges and universities face growing scrutiny over how they protect and manage data. Recent cyber incidents affecting major universities have heightened concerns about the resilience of educational technology systems and the operational impact of breaches.

Among education organisations, 98% expect AI infrastructure budgets either to rise or remain unchanged over the next year, with 46% anticipating an increase. Respondents also expect the share of AI projects generating positive returns to climb from 37% to 47% over the next 12 months.

These figures suggest institutions remain committed to AI spending despite mixed results so far. Many appear to expect stronger financial performance as projects mature, even as storage and access charges continue to weigh on budgets.

Andrew Smith, Director of Strategy and Market Intelligence at Wasabi, said the survey revealed tension between enthusiasm for AI and the cost of sustaining it.

"Education institutions are eager to dive head-first into AI, but the survey data illustrates a concerning trend regarding expectations vs. fiscal realities," Smith said.

"To ensure long-term success of AI initiatives, IT buyers in education must consider both the technical challenges associated with their data - including storage, migration and quality - and the long-term cost efficiency of accessing, retaining and securing it. Avoiding costly, budget-breaking fees from hyperscaler infrastructure services should be a priority."

Security concerns

The survey also highlights concerns about cyber resilience in the sector. Fewer than half of education respondents (47%) said they were completely confident in their ability to keep data unaltered and operational after a cyberattack.

Operational disruption remains a live issue. About 44% said they had lost access to public cloud data due to a cyberattack.

Even so, the research points to wider use of protective measures. Some 63% of education respondents said they now use immutability to protect data in the public cloud, up from 49% a year earlier.

That increase suggests institutions are taking more steps to prevent stored data from being changed or deleted after an attack. Yet confidence in recovery remains relatively low, indicating that investment in safeguards has not fully reassured technology leaders.

Survey scope

The research was conducted by Vanson Bourne for Wasabi and surveyed 1,700 IT decision-makers involved in public cloud storage purchasing across multiple sectors and countries. The education findings are based on 241 respondents from that wider sample.

The results offer a snapshot of a sector balancing ambitious AI plans with rising infrastructure costs and persistent cyber risk. For many institutions, the challenge is not whether to invest in AI, but how to manage the data storage and security demands that come with it.

In the education segment, 50% of respondents identified storage-related issues as the leading challenge in AI implementation, while 54% of cloud storage spending went on fees rather than core storage capacity.