[CKC] [May 2018] What subjects are most important for an AI safety researcher to know? (Open Call)

by Andrew Quinn1 min read18th May 20182 comments



[CKC]: I intend this to be a Community Knowledge Convergence project - that is to say, I do not think I am particularly well informed on the subject, and I think there's a lot to be gained from compiling an understanding of what others in the LessWrong community consider as very important, probably important, possibly important, or maybe important.

What academic subjects do you consider most important to the study of AI safety? What non-academic subjects would you recommend?

Currently, here are the subjects I think are most important.

  • (Non-academic) Eliezer's Sequences.
  • Mathematical logic.
  • Game theoretic economics.

You might be able to tell, I'm not very well versed in this discipline. Please comment and add more subjects for me to add to the [June 2018] posts.

2 comments, sorted by Highlighting new comments since Today at 9:45 AM
New Comment

I think Raemon has a good overview of everything here:

In addition, MIRI has a recommended textbooks list

Also, CHAI has a textbooks list here

Also, Toon Alfrink has started RAISE to help bring people up to speed.

(META: I think this topic has a very long history of already being discussed. See, for example, lukeprog on the best textbooks. Also, we've tried having monthly open-thread-like-discussion-things in the past, and they sort of flopped. So just a heads-up if you're aiming for something similar.)

Excellent resources! Thanks a ton LLL, I'll look through them in a few hours.