Writing AI prose that doesn't read like AI
Chapters7
Published7
Edition2026
Languageen-GB
The chapters.
In chronological order
Chapter 01
Why AI prose reads like AI
Why does AI writing sound flat? Next-token prediction averages toward the statistical middle. The structural cause and how to fix it, with real numbers.
Published
→
Chapter 02
The vocabulary tells (and why the blacklist isn't the fix)
The AI writing tells list a technical author audits against in 2026, why each phrase reads as generated, and why blacklist swaps just create new tells.
Published
→
Chapter 03
Rhythm and structural tells: why AI writing sounds robotic
AI writing sounds robotic when every paragraph runs the same length and pace. How to diagnose rhythm tells in a draft and fix the flat structure.
Published
→
Chapter 04
The draft starts in the second paragraph
How to edit an AI draft to sound human on a developer blog: delete the throat-clearing opener and start at the first sentence a reader can use.
Published
→
Chapter 05
Injecting first-hand experience into an AI draft
How to inject personal experience into an AI draft: an inventory of lived specifics swapped in for every generic claim the model produced.
Published
→
Chapter 06
You can't prompt your way to voice
How to keep your voice when writing with AI: voice prompts add surface markers, not register. What a style guide and a human edit do that prompts cannot.
Published
→
Chapter 07
A pre-publish style audit you can run
How to write a technical blog that doesn't sound like AI: a deterministic tells audit that blocks publish, and the two incidents that hardened it.
Published
→