Analysis cites filings-based concentration risk for Microsoft’s AI revenue tied to OpenAI

Microsoft’s AI push is drawing renewed scrutiny after a recent analysis argued that a significant share of the company’s AI business still runs through OpenAI.

The report, published by Yahoo Finance and focused on Microsoft filings, said that “around 70%” of Microsoft’s AI revenue is concentrated “entirely on OpenAI.” The framing underscores that even as Microsoft expands distribution of AI capabilities through cloud services, its monetization may still be tightly linked to a single external model provider.

Concentration risk is not a new theme for tech businesses, but it becomes sharper when the revenue base depends on upstream technology controlled by another company. If the analysis reflects the underlying revenue categories in Microsoft’s filings, it implies that Microsoft’s ability to meet customer demand for AI capabilities could be influenced by OpenAI’s roadmap, capacity, pricing, or availability.

The same Yahoo Finance piece characterizes the dependence as “somewhat enormous,” arguing the market may be underweighting the operational and commercial sensitivity embedded in Microsoft’s AI supply chain. That assessment is particularly relevant given how quickly enterprise customers are moving from pilots to production deployments and how often they evaluate vendor risk alongside performance.

Microsoft has not, in the material referenced by the report, detailed a clear diversification plan in quantitative terms that would offset the kind of dependency suggested by the filings-based estimate. The analysis therefore leaves open the question of how much Microsoft earns from AI features that rely on OpenAI models versus other approaches, including internal tooling and alternative model partnerships.

From a business perspective, this matters because AI revenue is typically not “one line item,” but a blended mix across cloud consumption, platform services, and product-level offerings. Concentration estimates can vary depending on what counts as “AI revenue” and which reporting definitions are used, so the key issue is whether the “around 70%” estimate maps consistently to how Microsoft reports and manages its business segments.

What remains unclear from the reported takeaway is the breakdown behind the estimate. The analysis points to a filings interpretation, but it does not, in the referenced post’s headline framing, provide granular numbers by product line, customer cohort, or contract structure, nor does it specify what proportion is tied to model usage versus other AI-related services.

Investors and enterprise buyers are likely to watch for more transparency in future filings and earnings materials, including any Microsoft disclosures that clarify how AI monetization is distributed across model providers and how it manages continuity risk if OpenAI terms change.