In partnership with
THE NUMBER: 12:1 — For every dollar the hyperscalers earn from AI, they're spending twelve building more capacity. $575 billion in capex this year. Alphabet just issued a century bond — the first by a tech company since Motorola in 1997. The debt matures in 2126. The chips it buys will be obsolete by 2029.
Anthropic now wins 70% of new enterprise deals. Not because of the Pentagon drama — because the product is better. Claude Code captured the developers, Cowork captured everyone else, and OpenAI's enterprise share dropped from 50% to 27% in two years. The morality play made great press. The product made the business.
But the company winning the product war can't fund its own infrastructure. Neither can OpenAI. Google and Meta can write the checks from operating cash flow. Everyone else is at the mercy of bond markets, investor sentiment, and a 12:1 spend-to-revenue ratio that requires 5x revenue growth in five years just to break even. If NVIDIA's annual chip cycle makes hardware obsolete in three years instead of five, that number jumps to 7.9x.
Protect online privacy from the very first click
Your digital footprint starts before you can even walk.
In today’s data economy, “free” inboxes from Google and Microsoft, like Gmail and Outlook, are funded by data collection. Emails can be analyzed to personalize ads, train algorithms, and build long-term behavioral profiles to sell to third-party data brokers.
From family updates, school registrations, medical reports, to financial service emails, social media accounts, job applications, a digital identity can take shape long before someone understands what privacy means.
Privacy shouldn’t begin when you’re old enough to manage your settings. It should be the default from the start.
Proton Mail takes a different approach: no ads, no tracking, no data profiling — just private communication by default. Because the next generation deserves technology that protects them, not profiles them.
We've watched this movie before. In the 1860s, railroads laid track as fast as they could borrow money. Then along came the automobile, the interstate highway, the shipping container — and the companies that had laid the most track right before that shift were finished. The AI equivalent of the car is already on the road. It's called specialization.
If you're sick, you don't want ChatGPT. You want the Harvard Medical School model — trained on proprietary clinical data by the best diagnostic minds in the world, drawing on datasets that never touched Reddit. The best maritime attorney in the world bills $3 million a year because hours are finite. But his knowledge isn't. Capture that judgment in a model and sell it to every shipping company, insurer, and port authority simultaneously — and you haven't automated a $3 million lawyer. You've created a $300 million judgment engine. The constraint that made expertise scarce just broke.
The highest-margin AI of the next decade won't come from bigger general models. It will come from specialized vertical intelligence running on a fraction of the compute. Which means $575 billion in general-purpose infrastructure might get repriced to utility returns — 8-12%, not 40% software margins.
Intelligence is becoming commodity. Judgment is not. The railroads are being built. The cars are already on the road.
-Harry & Anthony
On the site today: The full Signal/Noise briefing — Anthropic's 70% win rate, Tunguz's 12x math, the railroad-to-automobile pattern, and why the maritime lawyer's judgment engine is the highest-margin play in AI → getcoai.com
From the Scroll: Elon admits "xAI was not built right" and starts rebuilding → Read Full Story | Vinod Khosla predicts 80% of jobs done by AI by 2030 → Fortune