Knowledge

Short entries, each at most 256 characters of claim — with the measured numbers and sources behind it. No claim without proof.

2026-09-10 · ai · fundamentals

How AI works: an LLM is a next-word guesser at giant scale

An LLM is a next-word guesser at giant scale: text becomes numbers, flows through billions of learned weights, and out comes a probability for every possible next token. All it "knows" was pressed into those weights by predicting trillions of words.

evidence: 3 numbers · 4 sources →

2026-09-06 · tools · economics

Why tools that never upload can stay free

Web tools that compute in your browser cost their operator almost nothing per use — so they can stay free, without caps. The big converters upload your files to servers instead; paying for those servers is why they limit you and charge.

evidence: 4 numbers · 4 sources →

2026-09-06 · ai · discovery

llms.txt and AI bots: 97% of files got no fetches; coding agents fetch it, not by habit

AI search barely reads llms.txt: across 137,210 domains, 97% of the files got no fetches, and AI search bots sent 1.1% of the requests (Ahrefs, 2026). Coding agents do fetch it when a link points there. Publish one, link it, never build around it.

evidence: 7 numbers · 4 sources →

Technical fixes written for AI agents, the kind an assistant looks up when someone pastes an error, are in their own section.