1 day ago4 min read
Enterprise AI: Nothing to Fear, Plenty to Do
Why sensible companies buy enterprise AI backwards, told through one COO, four walls and a managing director who preferred interns.
top of page
The architecture of the AI-Native enterprise
Notes of EA
One argument, in order
Since July 2026 I've been publishing one long argument in weekly instalments: why the economy changed its question, why 95% of enterprise AI fails by design, what an LLM and an AI agent actually are, and what the enterprise must become. The posts and articles below are that argument in reading order. Start anywhere; the numbering will catch you.
1 day ago23 min read
The Wrong Search: Where Is the AI Money?
Why sensible companies buy enterprise AI backwards, told through one COO, four walls and a managing director who preferred interns.
Aug 242 min read
The revolution is becoming a framework
Each generation of AI models moves from a stochastic toy toward a controllable framework. That's not disappointment; it's when the technology becomes fit for real work inside a process.
Aug 242 min read
The AI bubble is being built for superstars. That is exactly why it will burst.
AI capabilities aren't overestimated; the way firms expect to convert them is. The parameter arms race benefits superstars, while profit remains in ordinary processes.
Aug 172 min read
What's your AI entropy budget?
System failure, process failure, human error: your risk register has three columns. Agents add a fourth: the error that arrives through normal operation. What's your entropy budget?
bottom of page