🐺 Microsoft just spent $37B in one quarter on AI (and Wall Street finally flinched)

Hey there,

Microsoft spent $37.5 billion on AI infrastructure last quarter. The stock dropped 5%.

That's the story this week. The S&P 500 crossed 7,000 for the first time, almost entirely on AI spending. DeepSeek published a paper on training efficiency that has analysts calling it a breakthrough. And Gartner now predicts 40% of enterprise apps will have AI agents by December—up from 5% today.

The money is flowing. The results are lagging. Here's what happened.

📰 WHAT'S TRENDING

Microsoft's AI Spending Spooks Wall Street Microsoft beat earnings expectations on nearly every metric. Revenue hit $81.3B, up 17%. Azure grew 39%. And then the stock dropped 5% after hours. Why? Capital expenditures hit $37.5B in a single quarter—up 66% year over year. CEO Satya Nadella said they added "nearly one gigawatt of total capacity this quarter alone." Investors are starting to wonder: when does all this spending turn into profit? 🔗 Read more

S&P 500 Crosses 7,000 for the First Time The index hit the 7,000 milestone on January 28. That's 14 months after crossing 6,000. The driver: AI infrastructure spending and pro-business fiscal policy. The data point nobody's talking about: 92% of U.S. GDP growth in H1 2025 came from AI-related spending. 🔗 Read more

DeepSeek Published a New Training Framework China's DeepSeek introduced "Manifold-Constrained Hyper-Connections" in a new paper. The goal: train bigger models for less compute. Counterpoint Research's principal analyst called it a "striking breakthrough." DeepSeek is preparing to release R2, its next flagship model. One year after DeepSeek's original "Sputnik moment," China's AI labs are still shipping. 🔗 Read more

Gartner: 40% of Enterprise Apps Will Have AI Agents by December Up from less than 5% today. 67% of business leaders say they'll maintain AI spending even if a recession hits this year. The blocker: 46% of organizations say integration with existing systems is their biggest challenge. 🔗 Read more

$100M+ Seed Rounds Are Now 40% of All Early-Stage Funding According to Crunchbase, over 40% of seed and Series A investment in 2026 has gone into rounds of $100 million or more. Recent examples: Humans& (a frontier AI lab) raised $480M seed. Ricursive Intelligence raised $300M Series A. Sam Altman's brain-computer interface company Merge Labs raised $252M seed. The bar for "startup" has officially moved. 🔗 Read more

Intel Stock Down 17% in January Alone Intel couldn't pivot its foundry business fast enough to compete with specialized AI silicon. Nvidia and AMD are capturing the AI hardware market. Intel is watching. The company that once dominated computing missed the platform shift. 🔗 Read more

Disney Characters Are Coming to OpenAI's Sora Disney and OpenAI signed a deal. Fans will be able to watch Sora-generated videos featuring Disney characters on Disney+. The partnership goes live in early 2026. Disney doesn't bet on technology platforms lightly. 🔗 Read more

🧠 HERE'S THE THING

Microsoft beat earnings expectations. Azure grew 39%. Revenue hit $81.3B.

Then the stock dropped 5%.

Why? Capital expenditures: $37.5 billion in a single quarter. Up 66% year over year. Capacity constraints through at least June.

Investors did the math. The infrastructure spending is outpacing the revenue it generates.

This is the largest infrastructure build-out since the internet. 92% of U.S. GDP growth in H1 2025 came from AI-related spending. That level of concentration is a bet, not a foundation.

Meanwhile, 95% of enterprise AI projects still aren't showing measurable ROI. Companies are buying the hardware before they know what to do with it.

The opportunity is in the gap between infrastructure and implementation. Someone has to help companies figure out what to do with all this expensive equipment.

💡 Smart Moves

Cold Truth: B2B Lead Gen That Works

If you run or work at a B2B company and you need more pipeline, I write a second newsletter called Cold Truth.

Twice a week. Tactical. Free thoughts. Cold email infrastructure, copywriting that gets replies, list building that actually works.

Everything is based on real campaigns we've run. Tap in if you’re running or working at a B2B business.

⏳ CONVERSION CORNER

Gartner says 40% of enterprise apps will have AI agents by December. But 46% of organizations say integration is their biggest barrier. That's the gap.

The companies winning right now build connective tissue—the middleware, the integrations, the "last mile" implementations that make AI actually work inside real businesses.

Now it's your move: If you're looking for an AI opportunity, don't compete on models. Compete on making existing models work with existing systems. That's where the demand is, and almost nobody is solving it well.

💎 DATA GEM

$37.5B That's what Microsoft spent on AI infrastructure in a single quarter—up 66% year over year.

For context, that's more than the GDP of over 100 countries. In one quarter. From one company. This is an arms race measured in gigawatts and billions.

That's it for this week. Stay sharp out there.

Jake

P.S. If this was useful, forward it to someone who's about to sign a seven-figure AI infrastructure contract. They might want the reality check first.