After going through Intro to ML Safety, I decided to condense the course into a ~30-page document. While I do not think this comes close to a substitute for the course, I figured this might be useful as a litmus test to see if someone would be interested in dedicating their time to self-studying ML/AI safety.

If it turns out that what I wrote misrepresents the course to a large degree/is just not thorough enough to be considered an introduction, I would be happy if you could point that out, and I may be inclined to write a second version!