According to The Information, Microsoft will pay to use Anthropic's AI technology to support Office 365 applications, marking the end of Microsoft's previous strategy of completely relying on OpenAI technology in its office suite. According to information from two informed sources, Anthropic's AI technology will work alongside OpenAI's technology to support new features in applications such as Word, Excel, Outlook, and PowerPoint.

This partnership reflects a subtle shift in Microsoft's relationship with OpenAI. Recently, OpenAI has been advancing its own infrastructure projects and plans to launch products that compete with LinkedIn, leading to some disagreements between the two companies. Microsoft's decision to partner with Anthropic is seen as an important move to diversify its AI partnerships.

Microsoft

According to reports, Microsoft's partnership with Anthropic was reached against the backdrop of negotiations between the company and OpenAI for a new agreement. OpenAI is currently restructuring its profitable operations, and Microsoft needs to ensure continued access to OpenAI's AI models after the restructuring. However, The Information pointed out that Microsoft's collaboration with Anthropic is not a negotiation tactic, but rather based on technical performance considerations.

According to sources, Microsoft's leadership believes that Anthropic's latest model, Claude Sonnet4, performs better than OpenAI's models in certain specific functions, especially in creating visually appealing PowerPoint presentations. This technical advantage became an important reason for Microsoft to choose Anthropic as a partner.

In fact, this is not the first time Microsoft has sought to diversify its AI technology suppliers. Although OpenAI remains Microsoft's default AI model provider, the company has already provided other AI models through the GitHub Copilot platform, including Grok from xAI and Claude from Anthropic. This multi-supplier strategy provides Microsoft with flexibility in choosing the most suitable AI technology for different application scenarios.

Microsoft is also actively advancing its own AI technology development. The company recently launched its first internally developed AI models, MAI-Voice-1 and MAI-1-preview, showing a strategic intention to achieve self-control in AI technology. Building internal R&D capabilities helps reduce reliance on external suppliers.

At the same time, OpenAI is also seeking to reduce its reliance on Microsoft. Last week, OpenAI launched a job posting platform that directly competes with Microsoft's LinkedIn. Additionally, according to a report by the Financial Times, OpenAI plans to collaborate with Broadcom to begin mass-producing its own designed AI chips in 2026. This means that OpenAI may run training and inference tasks on its own hardware in the future, no longer relying on Microsoft's Azure infrastructure.

Facing this change in the relationship, Microsoft spokesperson Michael Collins told TechCrunch that OpenAI will continue to be a partner for Microsoft in cutting-edge models, and the company remains committed to maintaining a long-term partnership. This statement shows that despite adjustments in the partnership, Microsoft still considers OpenAI an important strategic partner.

From an industry development perspective, Microsoft's partnership with Anthropic reflects the evolution of the AI industry's competitive landscape. As AI technology develops rapidly, major tech companies are seeking to establish a more diversified and stable AI supply chain to avoid risks associated with over-reliance on a single supplier.

This diversification strategy also has positive implications for enterprise users. By integrating the technological advantages of different AI suppliers, Microsoft can provide Office 365 users with more rich and specialized AI features. The complementary strengths of different models in specific tasks are expected to enhance the overall user experience and work efficiency.

Industry analysts believe that Microsoft's move marks an important shift in the cooperation model of tech giants in AI. From previous exclusive or primary reliance relationships, it is moving towards a more open and competitive cooperation model. This change may drive innovation and improve service quality throughout the AI industry.

Currently, Anthropic has not made an official comment on this collaboration. As the details of the cooperation become clearer, the specific impact of this strategic adjustment on Microsoft's Office product line and the entire enterprise-level AI application market remains to be observed.