Posts

Why Re-Reading Your Video Notes Doesn't Work

You open your notes from a course you took three months ago. You read a page. Most of it looks familiar — yes, that's right, you remember this. You close the file feeling fine about it. Two weeks later someone asks you what the course was about, and you can't say anything specific. Not a number, not an example, not one claim and who made it. The feeling is the problem Recognising something and being able to produce it are different abilities, and from the inside they feel almost identical. When you re-read your notes, every sentence arrives with a small click of "yes, I know this" — and that click is being generated by the page in front of you, not by your memory. Take the page away and it has nothing to stand on. Which is why the people complaining hardest about forgetting videos aren't beginners. They're the ones with the tagged Obsidian vault, the Notion database, the highlighted transcripts. The notes are good. The notes were never the problem. A...

How to Tell an AI Video Summary Is Fake (Three Checks, Ten Seconds)

Someone sends you a video. You paste the link into ChatGPT and ask what's in it. It answers straight away — clean, organised, confident. You skim it and move on. Here's how to check whether that answer was real. Three tells, ten seconds It's suspiciously even. Real talks ramble. Someone spends nine minutes on one point and thirty seconds on another. If every section in the summary gets roughly equal weight, you're reading an outline, not a talk. Nothing in it is specific. No numbers. No names. No "she argues that…". Just topics — the subjects that were covered, with nothing that was actually said about them. It matches the thumbnail. If the summary reads like a longer version of the promise on the thumbnail, that's because it is one. The single test that settles it: take one claim from the summary and ask whether it could only have come from someone speaking out loud. A number, a name, a moment where the argument turns. If nothing in the su...