cromulent2

Thank you. Investigative journalism is so important and I would happily believe some of the claims made here, but when I encounter even just a few sentences that sound LLM-written, suddenly I don't trust any of the statements in the source anymore. This site goes way beyond that, with a vibe-coded UI and generated articles. There might be value in what's reported here, but currently it requires a lot of work from the reader.

That makes me wonder: do you see a difference in when viewers drop off between using a more math-y title versus a more accessible one?

The "broke my brain" title originally put me off from watching. I caved after a few days; I think the video is one of your best!

Thanks! Frontend is Nextjs. Flask, Postgres and Redis in the backend. Hosted on a cheap VPS.

I spent way too much time trying to automate moderation (grouping related submissions, judging similarity, removing offensive submissions) using sentence-transformers and w2v. Finally gave up and settled on using deepseek to moderate for 0.002€ a day.

One thing I'm currently working on is improving the exploration vs exploitation balance in matchmaking (selecting the fakes to present to a player). I want sessions to contain good fakes, but every fake must be played a few times to judge its quality. As a first step, I'm replacing the ELO system for definitions with one that accounts for uncertainty (like Trueskill)