Articles
Notes from the engine.
Short, honest writing on consensus, building with AI and getting described correctly by the models people now ask.

AI video
Seven more AI experiments: every prompt, every take, one bug
Week two of one AI video a day: game covers, a street interview with AI people, a cookie-banner game, a tipping debate, anime, a song and the misses.
2026-11-15

AI video
Seven AI experiments in one week: every prompt and every failed take
One short AI video a day for a week, all in one workspace: pictures, a fake company, an ad, a song, an anime opening, a game and a model debate. Every prompt and every failed take.
2026-11-08

Code
Who Reviews the Code an AI Builder Writes?
All four AI app builders we track document an automated review of their own generated code. What that review covers, what it misses, and what it never asks.
2026-09-09

Idea validation
Can ChatGPT Validate Your Startup Idea?
One model's yes is tuned to please you. What sycophancy research found, why a single answer proves nothing, and how to make AI attack your idea instead.
2026-09-07

AI Visibility
How AI Assistants Describe Your Company, and How to Check
We ran the buyer questions against our own company twice, ten days apart. The method, the caveats, and why being described wrongly costs more than being absent.
2026-08-14

Consensus
Consensus vs a Single Model: What the Research Actually Shows
Does asking several AI models beat asking one? The measured answer, with sources: where consensus helps, where models agree wrongly, and when one is enough.
2026-08-12

Idea validation
How to Validate a Startup Idea Before You Build
A practical 2026 guide to validating a startup idea before you write a line of code: real failure data, why one AI opinion misleads, and what to test first.
2026-08-05