Microsoft reduces dependence on Openai by purchasing AI from rival humanity

Microsoft pays to use Anthropic’s AI in Office 365 apps, citing two sources. This move means that Anthropic’s technology will help drive new features in Word, Excel, Outlook and PowerPoint alongside Openai, and the end of Microsoft’s previous trust will be marked only for the productivity suite of ChatGPT manufacturers.

The movement to diversify Microsoft’s AI partnerships is pursuing its own infrastructure projects and potential LinkedIn competitors amid growing rifts with Openai.

Microsoft’s humanitarian deal comes as the company negotiates a new deal with Openai to ensure access to the AI ​​model after a pending commercial restructuring. However, the information says the move is not a negotiation tactic. Microsoft’s leaders say the latest model from Anthropic, the Claude Sonnet 4, performs better than Openai in certain features, specifically creating aesthetically pleasing PowerPoint presentations.

However, this is not the first time Microsoft has branched out. Openai is the default model, but Microsoft offers other models via Github Copilot, such as Grok from Xai and Claude from Anthropic. Microsoft is also trying to set itself up for independence. The company recently introduced its first two internal models, Mai-Voice-1 and Mai-1-Preview.

Meanwhile, Openai is similarly about to leave under the influence of Microsoft. Last week, Openai launched a job platform to take on Microsoft’s LinkedIn, and Financial Times reported that it plans to start mass production with its first AI chip in a partnership with Broadcom in 2026. This means that it may run hardware IT controls rather than relying on Microsoft’s Azure setup.

“As we said, Openai continues to be a partner in the frontier model and we are still committed to our long-term partnership,” Microsoft spokesman Michael Collins told TechCrunch.

TechCrunch contacted humanity for comments.

This article has been updated with comments and additional context from Microsoft.