Microsoft is broadening its artificial intelligence portfolio, agreeing to pay Anthropic for the use of its models in Office 365 apps, according to a report by The Information.
The move marks the first time the software giant is integrating Anthropic’s Claude models with OpenAI’s GPT technology across products such as Word, Excel, Outlook, and PowerPoint—apps that serve hundreds of millions of users globally.
The decision highlights both opportunity and tension in Microsoft’s AI strategy. For years, the company has leaned almost exclusively on OpenAI, into which it has invested more than $13 billion, securing a head start in the AI race. But developers found that Anthropic’s latest Claude Sonnet 4 model outperformed OpenAI’s GPT-5 in specific enterprise use cases, such as automating complex Excel financial functions or generating more visually polished PowerPoint presentations.
Register for Tekedia Mini-MBA edition 19 (Feb 9 – May 2, 2026): big discounts for early bird.
Tekedia AI in Business Masterclass opens registrations.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register for Tekedia AI Lab: From Technical Design to Deployment (next edition begins Jan 24 2026).
Microsoft will pay Amazon Web Services to access Anthropic’s models, a noteworthy arrangement given AWS is one of Anthropic’s largest shareholders and a direct competitor in the cloud market. Neither Anthropic nor OpenAI responded to requests for comment, while AWS declined to comment. Microsoft stressed it remains committed to OpenAI.
“As we’ve said, OpenAI will continue to be our partner on frontier models and we remain committed to our long-term partnership,” a spokesperson said.
The company said the price of AI features within Office 365 will remain unchanged despite the expanded technology stack. An official announcement is expected in the coming weeks.
A Web of Partnerships and Rivalries
The deal underscores how alliances in the AI sector are increasingly fluid. OpenAI itself has diversified beyond Microsoft, using CoreWeave, Google, and Oracle to meet surging demand for its models. Its ChatGPT assistant now reaches about 700 million people weekly, adding pressure on Microsoft to keep pace.
Microsoft, for its part, has acknowledged the competitive undertones. In its most recent annual report, the company listed OpenAI as a competitor for the first time—placing it alongside Amazon, Apple, Google, and Meta.
At the same time, Microsoft is investing in homegrown models. In late August, it began public testing of MAI-1-preview, an in-house AI model being evaluated on the benchmarking site LMArena. The company described the test as an early step toward enhancing its Copilot assistant for consumer use.
“We will be rolling MAI-1-preview out for certain text use cases within Copilot over the coming weeks to learn and improve from user feedback,” Microsoft said in a blog post.
The move complements Microsoft’s existing partnership work. At its Build 2025 developer conference, the company and GitHub announced they had joined the steering committee for the Model Context Protocol (MCP), an open standard pioneered by Anthropic that allows AI systems to better connect with external tools and data.
Mustafa Suleyman, CEO of Microsoft’s AI unit, struck an ambitious note. “We have big ambitions for where we go next—model advancements, an exciting roadmap of compute, and the chance to reach billions of people through Microsoft’s products,” he said in a post on X.
For now, Microsoft finds itself walking a delicate line that involves deepening its commitment to OpenAI while diversifying with Anthropic, investing in its own models, and leveraging standards like MCP.