Microsoft reportedly trains salespeople to disparage OpenAI and Anthropic

TechCrunch, citing Bloomberg, reports Microsoft is training its salesforce to talk down OpenAI and Anthropic. In a presentation, EVP Jacob Andreou reportedly compared Copilot directly against Anthropic's Claude, asserting that inside Microsoft's Office apps Claude was 'slower and less accurate, and lacked the proper security integrations.' The playbook reportedly targets both OpenAI and Anthropic.

The striking part is the conflict of interest: Microsoft is OpenAI's largest partner and investor and also resells Anthropic's models via Azure and Foundry, yet is now positioning Copilot against both. That reflects a strategic shift — Microsoft increasingly wants customers on its own first-party AI surface (Copilot, Foundry) rather than routing spend to partner models, especially as those partners become direct competitors in the enterprise.

The timing compounds developer frustration already simmering over Microsoft's pricing: the company is raising Foundry model deployment prices 10–50% in EU and non-US regions from September 1, which Azure-subreddit developers have branded a 'mandatory AI tax' tied to Copilot bundling. A sales playbook disparaging rivals lands as evidence, to critics, that Microsoft is leveraging its distribution to box out better models.

Competitively, the move is a tell that Copilot's raw model quality may lag Claude and GPT on some tasks — otherwise selling on integration and security rather than capability wouldn't be necessary. The community reaction (see pulse) questions the ethics given the partnerships. Watch whether OpenAI or Anthropic respond publicly, and whether enterprise buyers — who often run bake-offs — accept or reject the framing.