Anthropic and Microsoft Join Forces: Claude Comes to Microsoft 365 Copilot
Microsoft has announced that it is adding models from Anthropic to its Microsoft 365 Copilot tool. Specifically, these are Claude Sonnet 4 and Claude Opus 4.1, which businesses can now use alongside the existing options from OpenAI. This change primarily affects the Researcher agent, a tool for in-depth research, and Copilot Studio, where users can create their own agents to automate tasks. According to the announcement, this means greater flexibility for businesses, which can now choose a model based on their needs.
In the Researcher agent, Claude Opus 4.1 can now be used for complex tasks such as creating a detailed market entry strategy, analyzing product trends, or preparing quarterly reports. This agent searches not only the web and trusted third-party data, but also internal materials such as emails, chats, meetings, and files. Users simply select a model using a toggle, making it easy to switch between options.
How Does the Integration Work in Practice?
In Copilot Studio, Claude Sonnet 4 and Claude Opus 4.1 are available as options for creating and managing agents. This makes it possible to build systems for deep reasoning, workflow automation, and flexible tasks. For example, businesses can combine Anthropic models with those from OpenAI or the Azure catalog to tailor agents to specific requirements. According to Charles Lamanna, who leads Microsoft's business and industry applications team, this helps bring innovation from across the industry directly into workplace tools.
In an article on The Neuron, Grant Harvey describes how this integration works: users can see an option in the Researcher agent to switch between models with a single click. Claude offers strong capabilities in writing, synthesizing information from multiple sources, and automation, making it ideal for creating business documents or analyses. At the same time, he points out that when Claude is used, data is sent to Anthropic's servers, which means businesses must consider the applicable terms and regulatory compliance.
Benefits for Users
This collaboration between Microsoft and Anthropic gives millions of users access to Claude directly in Microsoft 365 Copilot. According to other reports, this reflects Microsoft's efforts to diversify its AI partners, especially at a time when its relationship with OpenAI is not developing in the best direction. Mustafa Suleyman, Microsoft's head of AI, mentioned in an interview that an organization as large as Microsoft needs maximum freedom of choice.
Claude Opus 4.1 excels at tasks requiring deep reasoning over complex data, while Sonnet 4 is suited to faster automation. Businesses can therefore choose based on whether they need nuanced writing, intelligence for technical tasks, or a long context window for tracking details. All of this is available without having to switch between platforms—users simply select an option from the menu.
How to Get Started and What to Consider
To get started, a company administrator must enable access to Anthropic models in the Microsoft 365 admin center. This feature is being rolled out through the Frontier Program for customers with a Microsoft 365 Copilot license. According to the instructions on Microsoft's website, this is done in the Copilot settings, where the AI provider for Researcher and Copilot Studio is selected.
It is important to be aware that when Claude is used, data is processed outside Microsoft's environment under Anthropic's terms. This means that Microsoft's standard guarantees regarding compliance, audits, and data residency do not apply—instead, Anthropic's rules apply. Businesses should review these terms to ensure the security of their information.
This integration is only the beginning, as Microsoft states in its blog. More experiences with Claude in the Microsoft 365 ecosystem are expected soon, promising even more options for businesses seeking reliable AI tools.
Source: anthropic.com
