The Human-AI Relationship — Stone’s Notes from Practice

First entry: 2026-10-08 · 中文版

---

Thesis (2026-10-13)

Direction: work with AI — and be better at it than everyone else.

Three things, unpacked: clear intent — know what you want, so you can spot when the AI is wrong; a verification system — cross-check across models + test it yourself, never get led by the nose; taste — knowing what’s actually good versus what merely looks right. Multiply the three together and you get agency × taste.

---

Contents

Part One: Foundations — the four starting points beneath the thesis, plus Muse’s first review.

• 1. AI won’t replace humans, but it will replace human grunt work
• 1. AI’s “creativity” isn’t true creativity — it’s just sentence-meaning you haven’t seen before
• 1. The migration of accomplishment: from “writing it myself” to “steering AI to follow my intent”
• 1. AI makes mistakes: in familiar territory you can catch them; in unfamiliar territory, use another AI to review
• Muse’s review (2026-10-08)

Part Two: Working with AI — Behaviors and Methods — AI’s habits, and how to deal with it.

• 1. Notes on AI coding: the code looks perfect but probably has bugs; human testing and feedback are irreplaceable
• 1. Drive the company’s AI like it’s stolen: joyride in the playground, be an engineer in production
• 1. AI has no “memory,” only context: check the harness first, blame the model second
• 1. What makes a magic prompt work isn’t the incantation — it’s the principle
• 1. Understanding debt and cognitive offloading: never let go of the steering wheel
• 1. AI slacks off — humans have to play foreman
• 1. AI’s three “human traits”: over-eager, lying, and lazy

Part Three: The Human Position — Value, Craft, and Ambition — where human value lies, how to keep your craft, how big to dream.

• 1. Quantity without quality is real — but veterans get both: agency × taste
• 1. Three things AI can’t replace: intuition, taste, and veto power
• 1. AI can’t catch the jokes you never said out loud: it’s missing context, not knowledge
• 1. With the grunt work gone, how do young people learn the craft?
• 1. The programmer as a grandmaster playing five boards at once: placing pieces can be outsourced, reading the board cannot
• 1. Where does my ambition sit? Three suggestions

Part Four: Outside Input — Cases and Readings — cases and articles I’ve read, distilled, with source links.

• 1. The neo-feudal layering of the AI era (one person’s take: pretty framework, dubious predictions)
• 1. Ambition in four measures: Rieseberg’s driver’s manual

---

To be continued

New entries go into the matching volume by theme; numbering continues globally. More to come.