Microsoft’s AI moat is still real, but it no longer looks exclusive
Microsoft still looks like the best-distributed enterprise AI platform, but the market is no longer treating Azure as the only road through OpenAI. That shift matters now because OpenAI’s multi-cloud turn and rising capex scrutiny widen the re-rating debate to Alphabet, Oracle, and even infrastructure beneficiaries like CoreWeave.

The cleanest way to read Microsoft right now is not as a broken AI winner, but as a winner whose monopoly narrative is fading. The old market shortcut was simple: if OpenAI kept scaling, Azure would be the primary toll collector. That shortcut got weaker the moment OpenAI’s infrastructure story became visibly multi-cloud, with Google Cloud now part of the supply mix after Azure had been the exclusive provider until January 2025. Microsoft’s moat is still real because distribution, enterprise relationships, and product integration still matter more than raw model access alone — but that is a different thesis, and it deserves a different valuation debate.
That distinction matters because Microsoft’s own numbers already support a reframing. Azure grew 40% year over year in fiscal Q3 2026, which is still elite by any large-cap standard and hardly evidence of a collapsing lead. But the market is also looking at the quality of that growth differently: commercial bookings excluding OpenAI grew 7%, while headline bookings were distorted by lapping a large OpenAI commitment. In other words, Microsoft is still winning AI, but the OpenAI piece is no longer a clean stand-in for Azure’s entire future. Once investors stop treating OpenAI demand as Azure-exclusive,


