Microsoft, Meta Push Staff Away From Anthropic's Claude To In-House AI Tool

Microsoft and Meta have each scaled back internal use of Anthropic's Claude by steering employees toward their own AI products. The two companies are among Anthropic's biggest corporate customers, which makes any pullback a challenge for the AI firm's efforts to keep its rapid growth going.

Microsoft had been on course to spend $1 billion this year on internal use of Anthropic's technology, a report by The Information said. That projection has since been cut by a third. The spending covers Claude Code, Claude models offered inside Microsoft's Copilot, and Claude Mythos.

A Microsoft spokesperson reportedly said the company has been directing staff to its GitHub Copilot coding tool, though engineers remain free to pick other models.

At Meta, the number of employees using Claude Code has fallen from roughly 60,000 earlier this year to about 30,000, according to the report. The company got there mostly by nudging staff to Muse Code, its rival to Claude Code, and MetaCode, a tool built for internal use only. Meta had been headed for billions of dollars in AI tool spending this year, with Claude Code its most popular tool.

The shift matters for Anthropic because both companies rank among its largest corporate customers, and the report frames the pullback as a challenge to the AI firm's efforts to sustain its rapid growth.

It also comes as companies struggle to predict what AI will cost them. WSJ reported on Monday that only 11 percent of nearly 400 businesses in a recent survey could accurately forecast their AI spending. The Journal noted that AI behaves less like conventional software and more like a human worker, since it takes actions and makes decisions and errors as it goes.

Anthropic remains the leader in business adoption. According to the Ramp AI Index, 43.5 percent of US businesses were paying for its subscriptions or tokens as of July. The index also showed how uneven AI spending is. In July, the top 1 percent of US businesses spent a median of $7,400 per employee on AI, against $11.95 for the median company.