Enterprise Software Waste Drains 28% of Budgets, Starving AI Investment

A new survey of 2,018 chief information officers and senior IT decision‑makers at companies with 10,000+ employees and $1 billion+ in annual revenue shows that enterprises allocate an average of 28 percent of their software budgets to upgrades, migrations, and licensing changes they did not request and often believe will degrade performance or reliability.

Ninety‑five percent of respondents said their organization had paid for at least one vendor‑driven change in the past three years that they expected would reduce business value, and 53 percent had done so repeatedly. Seventy‑six percent reported having very little negotiating leverage, while 80 percent warned that switching away from major vendors would cause significant operational disruption.

Why it matters for GPU and AI infrastructure

The budget drain hits innovation hardest: 36 percent of leaders said unwanted upgrades divert funds from new initiatives, 32 percent cited human‑resources impacts, 31 percent pointed to security, and another 31 percent noted reduced customer‑facing investment. For organizations building AI capabilities, every dollar locked into forced software cycles is a dollar not spent on GPU compute, model training, or specialized AI hardware — areas where AiGpu’s flexible cloud platform helps teams avoid vendor lock‑in and align spend with actual workload needs.

Liberty Mutual’s CIO highlighted that a single upgrade cycle can occupy top engineers for the better part of a year, creating a recurring drag on the very talent needed to differentiate through AI. Ninety‑one percent of surveyed leaders support regulatory protections comparable to long‑term residential tenancy, signaling broad demand for contractual fairness that would also benefit GPU and AI infrastructure procurement.