I don’t think it is relatively difficult to make “Ethical” AI.

Simply refer to the sources you used and make everything, from the data used, the models and the weights, of public domain.

It baffles me as to why they don’t, wouldn’t it just be much simpler?

Except tox’s graphical clients aren’t maintained anymore

How is bandcamp involved here? I’m curious now

There’s this project: https://watchit.movie/#/ https://github.com/ZorrillosDev/watchit-app

I never got to try it, but they are doing exactly what you want it seems. I don’t know how much of their backend they have open-sourced, though. I definitely need to research this more.

They do have a matrix channel and a fediverse presence, so they already look quite alright from the start, and the project is very interesting.