Microsoft has instructed employees across Windows, Teams, M365 and other divisions to install and test Claude Code alongside Copilot, with the tool reportedly approved for use on all Microsoft repositories.
What’s reported
- Internal testing of Claude Code is said to be carried out across multiple product groups, running in parallel with Microsoft’s Copilot.
- The company’s broader commercial posture includes a $13 billion stake in OpenAI while reportedly spending $500 million per year with Anthropic.
- Azure sales teams are reportedly receiving quota credit for selling Anthropic services, suggesting a commercial incentive to promote Anthropic offerings alongside Microsoft’s own AI products.
Why this matters
These details point to a multi-vendor approach within Microsoft’s AI tooling strategy: integrating or validating alternatives alongside in-house and partner solutions. For engineering teams and repository governance, the note that Claude Code is allowed on Microsoft repos is a concrete policy signal about internal experimentation and toolchain flexibility. The reported spending and sales incentives also highlight how commercial relationships with multiple AI providers may be shaping product and sales behavior.
Next steps for readers
The brief report provides a snapshot of internal activity and vendor commitments rather than extensive technical analysis. For those interested in the source report and community discussion, the original post is linked below.
Original source: https://www.reddit.com/r/ClaudeAI/comments/1qk4up5/microsoft_is_using_claude_code_internally_while/?