Microsoft made $10 billion in revenue from OpenAI in the three months ending September 30. That’s not profit, but it’s still a staggering number that shows just how deeply intertwined the two companies have become.

The figure, disclosed in Microsoft’s Q1 fiscal 2027 earnings, includes revenue-sharing payments and other commercial arrangements stemming from Microsoft’s investment in and partnership with OpenAI. It’s the first time we’ve seen the relationship quantified at this scale, and it underscores something important: Microsoft isn’t just OpenAI’s backer. It’s also OpenAI’s largest customer, cloud provider, and distribution partner.

This matters because Microsoft just posted its tenth consecutive quarter of record revenue. The company brought in $92.3 billion in Q1, up 19 percent year-over-year, with net income jumping 32 percent to $33.6 billion. And for the first time, Microsoft disclosed Azure’s actual quarterly revenue in dollars: $33.2 billion. That puts it within striking distance of AWS, which has long dominated the cloud infrastructure market.

The OpenAI revenue is part of a larger story about how AI is reshaping Big Tech’s financials. Microsoft restructured its entire financial reporting this quarter, splitting its businesses into two segments instead of three. The change makes it easier to see how much of the company’s growth is coming from cloud and AI services versus legacy software.

Google posts highest search growth in five years

Google parent Alphabet also reported Q3 earnings this week, and the numbers tell a similar story. Revenue hit $127.6 billion, up 25 percent year-over-year. CEO Sundar Pichai said AI investments are driving search query growth to its highest rate in five years, particularly for commercial queries.

“Our AI investments are driving overall Search query growth, including commercial queries,” Pichai said in a statement. “In fact, our year-over-year query growth rate is the highest it’s been in the last five years.”

That’s notable because it suggests AI Overviews and other generative features aren’t cannibalizing traditional search traffic. Instead, they appear to be bringing users back to Google for queries they might have skipped or taken elsewhere. The company is also expanding AI-powered search ads globally as core advertising growth slows.

Google released another frontier model this week, Gemini 4 Carbon, which the company says offers “significant gains” in coding, agentic tasks, and cybersecurity. It’s currently available only to select cybersecurity defense customers, but Google highlighted Carbon’s top spot on the Artificial Analysis Intelligence Index benchmark, ahead of Claude Opus 5.5 and GPT-6 Astra.

The infrastructure buildout continues

What ties these earnings together is infrastructure spending. Microsoft, Google, and Amazon are all pouring billions into data centers, chips, and the physical backbone needed to train and run AI models. That spending is now showing up as revenue growth across the board.

Microsoft’s Azure revenue reflects not just OpenAI’s usage but also enterprise customers building their own AI applications on top of Azure’s infrastructure. Google’s search growth is driven by users interacting with AI features that require more compute per query. The companies are betting that AI will expand the total addressable market for cloud services, not just shift usage patterns.

Anthropic, meanwhile, launched new cost analytics tools this week to help companies understand what their AI spending is actually buying. The timing isn’t coincidental. The startup is reportedly planning an IPO, and transparency around usage and costs is becoming a competitive differentiator as enterprises try to figure out whether these models are worth the money.

The other thread running through this week’s news is the growing divide between companies that can afford to build AI infrastructure at scale and everyone else. Microsoft disclosed Azure’s revenue for the first time because the numbers are good. Google highlighted query growth because it’s winning. Anduril and Qualcomm announced plans to build laptops entirely in the United States, starting in 2027, because defense contracts and domestic manufacturing requirements are creating new opportunities for companies willing to take on the capital expense.

None of this is cheap. Microsoft’s capital expenditures remain astronomical, and Google’s margin compression reflects the cost of running AI-enhanced search at scale. But the bet appears to be paying off. The question now is how long this growth rate can last, and whether the models themselves will become commoditized before the infrastructure investment pays back.