Corporate America Rations AI as Soaring Costs Force Shift to Cheaper Tools

A new report indicates that Corporate America is beginning to feel the sting of the AI boom's massive price tag. Leading companies are starting to ration AI usage as the cost of computing tokens and specialized software licenses exceeds initial budget projections. In one notable example, ride-hailing giant Uber reportedly exhausted its entire annual AI budget by March, forcing a re-evaluation of how and where the technology is deployed.

The skyrocketing costs are driven by the intense demand for high-end GPUs and the premium pricing models adopted by providers like OpenAI, Microsoft, and Google. As a result, businesses are increasingly looking for 'small language models' or open-source alternatives that provide 80% of the functionality at a fraction of the cost. The era of 'limitless AI experimentation' appears to be giving way to a more disciplined, ROI-focused approach.

To help clients navigate this rocky transition, Big Four accounting firms like KPMG are aggressively scouting Silicon Valley startups. KPMG is retooling its own business by training staff to build AI agents and trying to sell AI advisory services to help clients manage these soaring costs. The goal is to prevent 'AI disruption' from eroding margins while helping corporations find efficient ways to integrate the technology.

This shift toward cost-consciousness is creating a second life for older tech players. Rackspace, backed by Apollo Global Management, is finding new relevance by helping companies manage private cloud workloads for AI. By moving AI tasks out of the expensive public clouds and into managed private environments, companies like Rackspace are positioning themselves as the 'budget-friendly' alternative for enterprises that need AI power without the runaway costs of the major hyperscalers.