Co-Intelligence

Two of the ideas you’ll keep

The Clif summary of Co-Intelligence pulls out 7 insights. Here are the first two.

1. Treat AI as a person, not software: Mollick's advice to 'imagine it as a human' isn't literal, it's a shortcut. LLMs respond better to conversational instruction, role-assignment, and iterative feedback than to rigid commands, because they were trained on human output.
2. The Jagged Frontier means AI's ability is wildly uneven and invisible from the outside. It writes a brilliant sonnet but botches simple arithmetic, so the only way to map what it's good at is to use it constantly for your own tasks.

The other 5 — and the full 11-minute summary — are in the app.

What’s inside: 9 chapters

The whole summary runs about 11 minutes, read or listened to.

1. Sleepless Nights and an Alien Mind
2. Principle One: Always Invite AI to the Table
3. Principle Two: Be the Human in the Loop
4. The Jagged Frontier and the BCG Experiment
5. Principle Three: Treat AI Like a Person (But Tell It What Kind)
6. Principle Four: Assume This Is the Worst AI You'll Ever Use
7. Cyborgs, Centaurs, and Reinventing Work
8. AI as Tutor, Coach, and Creative Partner
9. Four Futures and Living With the Alien