A growing number of tech companies are beginning to question whether the massive cost of generative AI is truly sustainable, even as the industry continues to promote artificial intelligence as the future of work.
A viral post on X this week claimed that “Amazon has reportedly scrapped its internal AI leaderboard as costs soared,” adding that a senior executive told employees, “Don’t use AI just for the sake of using AI.”
Another post listed several examples of companies allegedly struggling to control AI-related expenses. It claimed that one company spent $500 million on Claude in a single month because no usage limits had been set.
The post also alleged that Uber had created leaderboards ranking engineers based on how much AI they used, rather than what they actually delivered. It further claimed that Uber had exhausted its entire 2026 AI budget by April, with its COO saying he could not connect the spending to any clear consumer-facing features.
The same post said that a CTO told Axios that some employees were using enterprise AI tools simply to check the weather. It also claimed that Microsoft had cancelled many Claude Code licences because the token bill had spiralled out of control.
The post ended with a sharp warning about the pressure now facing companies: “Companies are now laying people off to pay the AI bill. Not because AI replaced the work. Because the bill replaced the headcount.”
According to The Verge, Microsoft has reduced the use of Anthropic’s Claude Code licences for many employees, with cost reportedly playing a role in the decision.
In a recent interview on the “Rapid Response” podcast, Uber executive Andrew Macdonald acknowledged the growing disconnect between rising AI spending and visible consumer benefits. His comments added to concerns already spreading across the tech industry.
Macdonald said it had become difficult to clearly link Uber’s expanded use of Anthropic’s Claude Code tools with innovations that directly improved customer experience.
The comments reflect a broader trend emerging across Silicon Valley. Companies that once aggressively pushed AI adoption are now reassessing spending after operational expenses rose faster than expected.
Several media reports suggest that internal restrictions, cancelled AI licences and warnings over uncontrolled AI usage are becoming more common as firms try to bring costs under control.
Last month, Axios senior AI correspondent Madison Mills told CNN that she had been hearing directly from companies about the scale of their AI spending.
For years, the belief that AI would eventually replace millions of workers helped fuel massive investment across the technology sector. The promise of automation became one of the biggest arguments behind the AI boom.
But as operating costs continue to rise, some companies are now facing a different reality: keeping human employees may still be cheaper than running large-scale AI systems.
