Meta and Microsoft take steps to reduce employee usage of Claude AI
According to an October 5 report by The Information, Microsoft had projected spending at least $1 billion on Anthropic technology for its own internal use this year, but later reduced that estimate by more than a third. Within its cloud and AI division, monthly AI spending limits reportedly dropped from $100,000 per employee to around $10,000 in most cases. At Meta, layoffs explain part of the drop in Claude Code users, but the primary driver is Meta's push to develop its own AI tools. MetaCode had more than 30,000 users and Muse Code more than 6,000. Even so, Meta spent more than $105M in a recent 28-day period on Claude Code. The cuts only affect internal use. Customer spending on Anthropic models through Microsoft's enterprise platforms continues to grow. Anthropic's IPO prospectus said two customers represented approximately 25% of its revenue, though it did not name them. For founders, the lesson is about customer concentration. Large buyers that also compete with you can move token spend to their own tools fast once costs climb, so heavy usage from them is no guarantee the revenue will last.
