Some companies in 2026 are spending more on artificial intelligence infrastructure and compute power than on employees’ salaries. Bryan Catanzaro, Nvidia’s vice president of applied deep learning, said, “For my team, the cost of compute is far beyond the costs of the employees” [1]. This reflects a shift in how tech firms allocate resources amid growing AI workloads.
Uber’s chief technology officer revealed the company exhausted its entire AI budget for 2026 due to token usage costs alone, highlighting how expensive running AI models can be [1]. Meanwhile, Amos Bar-Joseph, CEO of Swan AI, said his company is expanding by leveraging AI spending instead of adding staff. “We’re building the first autonomous business—scaling with intelligence, not headcount,” he said [1].
Worldwide IT spending is forecast to reach $6.31 trillion in 2026, a 13.5% increase from 2025. The growth is driven primarily by investment in AI infrastructure, software, and cloud services, according to industry projections [1]. This surge places new pressure on companies to demonstrate clear returns on AI investments through productivity gains or other metrics.
Brad Owens, vice president of digital labor strategy at Asymbl, said companies are increasingly questioning the true value of human versus digital labor, noting, “The tone is shifting a bit more into what is the true value of a worker... human or digital?” [1]. This reassessment comes as some AI labs have raised prices to handle surging demand, potentially turning AI spending from a competitive edge into a financial burden [1].
OpenAI investors have suggested Codex is more efficient in token usage than Anthropic’s Claude Code, while Anthropic itself recently updated pricing plans to manage increasing customer demand [1]. These adjustments underline the tension between rising AI compute costs and the need for cost-effective AI deployment.
The growing AI infrastructure expenses and their impact on company budgets will remain under close watch throughout 2026, as IT leaders juggle rising costs with pressures to scale AI capabilities.