Microsoft's AI spending is producing faster growth with better economics: Azure accelerated to 48% on a $100B+ base while capex was $39.4B, below expectations. Management then guided Azure to 49-50% and said demand still exceeds supply, ending the debate over whether AI investment is paying off. Overweight, $625 target. x.com  x.com  x.com 

Microsoft's Azure has lost its OpenAI-model access edge because those models are also on AWS and are coming to Google Cloud. Azure now has to win on price and service rather than access. x.com 

Zhipu AI stands to gain as AI models commoditize and results hinge on engineering, where Chinese companies are strong. A model's effectiveness is model times engineering times compute; China cannot yet beat GPT's base models, but engineering in coding is becoming more important and masks model advantages. As models commoditize, Chinese models catching up through engineering and reputation is inevitable, and a breakthrough "Seedance moment" in coding could come soon. x.com  x.com  x.com  x.com 

Microsoft's 50% Azure growth and prospect of positive cash flow in 2027 make the stock look bulletproof. Headcount cuts, lower Anthropic spend, a better Copilot routed to cheaper tokens, and coming proprietary chips and model usage add to the case. It should get Apple's multiple plus recognition for cybersecurity. x.com  x.com 

Hyperscale cloud providers are being revalued as more than compute resellers, as open-source and in-house mid-size models reshape their costs and could lift gross margins. Their Q3 results in mid-to-late October could be very impressive. Per Barclays, $35–40 of every $100 AI labs earn flows to AWS, Azure and Google Cloud as inference spend, at 35–45% operating margins, with stable share over two years. Rising long-term yields cap gains until the 10-year Treasury falls below 4.6%. x.com  x.com 

Zhipu AI is a Chinese AI arbitrage, shipping models near frontier Western performance at a fraction of the valuation. Its GLM-5.1 scored 45.3 on a coding benchmark versus 47.9 for Claude Opus 4.6, a 2.6-point gap far smaller than the valuation gap. Zhipu is up 357% since its January IPO, double MiniMax's return, yet both trade at about $40B. x.com 

Microsoft, Alphabet, Amazon, Oracle, Meta and SpaceX will be long-term AI winners because value is shifting from commoditizing models to the ecosystems around them. With gaps closing, costs plunging and easy switching, models lose pricing power while data, tools, inference infrastructure and apps capture the upside. Monetization comes from bundling models into massive distribution, like Gemini across Search and Android; model makers will thrive only by partnering with ecosystems. x.com  x.com 

AI agents' productivity over human labor should lift GPU demand and rates for neoclouds. The bottom looks in for CoreWeave, Nebius, Boost Run, Sharon AI and IREN. Gavin Baker says token prices can fall as revenue per GPU rises; paraphrasing Elon, even $38 per GPU-hour can pay off. CoreWeave's five-year fixed contracts may lag current rates; Boost Run has less capacity locked in. If SpaceX takes more growth, slower cash burn could rerate neoclouds, though smaller customers may pay less. x.com  … Show more

Microsoft is building its enterprise AI platform to be model-agnostic, keeping the harness, context, memory and action space separate so any model is swappable. Customers want to pick models by quality, latency, cost and regulation, and the number using multiple providers has risen fivefold this year. Satya Nadella's commitment to that design matters, and OpenAI and Anthropic are unlikely to appreciate it. x.com 

Open models still need GPUs, memory, networking and power, so cheaper intelligence can bring more workloads to hyperscalers' paid infrastructure. Gavin Baker notes that OpenAI and Anthropic's revenue misses parts of the AI economy. BEP's OpenRouter chart shows rising token use for selected open models; AWS, Microsoft and Google Cloud are illustrative beneficiaries, not verified hosts of those volumes. x.com