Microsoft Corporation has begun a large-scale transition to its own artificial intelligence models, MAI (Microsoft AI), in Excel and Outlook applications. This move marks a strategic departure from the costly use of third-party solutions from OpenAI and Anthropic and aims to significantly optimize expenses.
Currently, tens of thousands of weekly queries in these office applications are already being processed exclusively by Microsoft's internal systems. The company, which previously had access to OpenAI's computing power on favorable terms due to a long-term partnership, now seeks to reduce dependence on external providers and take control of pricing.
Why is Microsoft changing its provider?
The key reason is cost efficiency. The AI division, led by Mustafa Suleyman, is unwilling to accept the prospect of rising costs after the cancellation of discounts previously offered by market leaders. According to my data, until now, Excel and Outlook relied heavily on technologies from OpenAI and Anthropic. However, the share of MAI models in the total volume of AI tasks is still small, indicating a gradual but steady migration process.
As early as June, Suleyman directly stated the intention to reduce spending on Anthropic and fully transition to in-house developments. "We pay Anthropic a lot of money—our task is to gradually reduce costs and completely eliminate unnecessary expenses," he emphasized. This decision is part of a broader strategy to optimize the budget. As early as May 2026, Microsoft began phasing out most internal licenses for Claude Code.
The shift to proprietary models is just one element of a large-scale cost-cutting effort. Concurrently, the company announced layoffs of 2.1% of its workforce, amounting to 4,800 people. The main impact will be on the Xbox gaming division, where approximately 3,200 specialists will lose their jobs.
Cryptalist Analysis: This move by Microsoft is a classic example of vertical integration in the AI era. Control over proprietary models not only reduces operational costs in the long term but also provides technological sovereignty. For the cryptocurrency and blockchain project market, which is increasingly integrating AI, this is a signal: relying on a single vendor is risky. Diversification and in-house development are key to resilience in a rapidly changing landscape.