Microsoft is reducing its reliance on OpenAI by integrating AI technology from competitor Anthropic into its Office 365 suite, according to sources cited by The Information.
Anthropic’s models will power new features across Word, Excel, Outlook, and PowerPoint, marking a shift from Microsoft’s previous exclusive use of OpenAI.
This move comes amid growing tensions with OpenAI, which is developing its own infrastructure and launching an AI-driven jobs platform to rival Microsoft’s LinkedIn.
OpenAI also plans to produce proprietary AI chips with Broadcom in 2026, enabling it to run AI training and inference independently from Microsoft’s Azure cloud.
While OpenAI remains a key partner, Microsoft is expanding its AI options, also offering models like xAI’s Grok and Anthropic’s Claude via GitHub Copilot. Microsoft believes Anthropic’s Claude Sonnet 4 outperforms OpenAI in certain tasks, such as creating polished PowerPoint presentations.
The company is simultaneously advancing its own AI models, including MAI-Voice-1 and MAI-1-preview, indicating a push for greater self-reliance.
Microsoft and OpenAI are negotiating ongoing access to OpenAI’s technologies amid OpenAI’s planned restructuring.
“As we’ve said, OpenAI will continue to be our partner on frontier models and we remain committed to our long-term partnership,” said Michael Collins, a Microsoft spokesperson.