Crash goes MSFT


If you happen to have stock in Microsoft (MSFT), you've perhaps noticed a certain underperformance in recent weeks. Well, months, but had been a pullback from an all-time high near \$550/share to the low \$500s has turned into a nosedive touching $400 today. This on the backs of their most recent quarterly reports beating estimates and other AI stocks holding their own.

The common wisdom is that the crash is due to Microsoft's Azure growth, while beating the official estimate, was ever-so-slightly less than Wall Street's internal whispering (estimate: 38%, reality: \$39%, whisper: \$41) while Microsoft's \$600B+ backlog is dominated by promised revenue from OpenAI and requires even more capital expenditures on AI datacenters with a limited lifespan.

By that common wisdom, one would think that MSFT is a buy. Or maybe it's the canary in the coalmine for the entire "AI Bubble" (I put Bubble in quotes because, like the 2000 dot-com bubble, it's more a matter of timing than of "no, that's not going to matter.") Or, I guess it needs to be acknowledged, it could be that the market has decided that OpenAI, long seen as the front-runner in the LLM space, has stumbled in the backstretch.

Google's Gemini 3 has undoubtedly leapfrogged the other LLMs in terms of general "chat" quality and Anthropic has pivoted towards domain-specific monetization in the form of Claude Code (apparently hitting $1B annual revenue) and some just-announced legal and medical fine-tuned models. OpenAI, on the other hand, underwhelmed with ChatGPT 5. ChatGPT 5.3 is expected any minute now (like, I wouldn't be surprised if it hits before I post this), but perhaps the market thinks that it won't keep pace with other advancements.

Honestly, I have no idea. In recent months I cycled off ChatGPT to Gemini as my daily driver for chats, but have stayed with OpenAI's Codex until just the other day, when it became Claude Code's turn to be my software development assistant.

I imagine that a big part of MSFT's decline is stop-loss and squeeze selling and that, gee whiz any moment now, it'll bounce back up by dozens of dollars per share. But maybe not. At some point, the market is going to decide on the horse race and there will be winners and losers. It's hard to imagine Microsoft not being one of the win-place-show oligopoly, but who knows?

It's worth pointing out, I think, that simultaneous with this MSFT crash, there has been an absolute frenzy around the release of OpenClaw (nee MoltBot nee ClawdBot), a personal agentic gateway. A frenzy so strong that it is visible in Apple's supply line for Mac Minis and the stock price of Cloudflare. And Anthropic's release of a legal product apparently sent the prices of some legal services companies down. These are just releases and they are moving the market.

There's a there there.