Meta and Microsoft cut back internal Claude use
The Information says two major Anthropic customers are steering staff toward in-house AI tools, pressuring Claude adoption inside big tech.

Meta and Microsoft are working to reduce employees’ reliance on Anthropic’s Claude, according to a report by The Information that was also picked up by Yahoo Finance, Investing.com and TheStreet.
The report says Microsoft lowered projected internal spending on Anthropic technology by more than a third as it pushes staff toward its own AI systems. It also says Meta’s Claude Code user count fell to about 30,000 from roughly 60,000 earlier in the year.
The details are notable because both companies have been important enterprise customers for Anthropic, even while building competing AI products. If large technology companies can shift more coding, productivity and internal-assistant work to their own models, it could put pressure on paid usage growth for outside AI labs.
For readers, the takeaway is not that Claude is disappearing from big tech workflows. It is that the enterprise AI market is becoming more self-reliant: the largest customers increasingly want to own the tools, model routing and spend controls behind everyday AI work.



