I’m a Ph.D student at MIT working in AI safety--mostly interpretability in deep networks. I thought I would very briefly share a list of links that I have bookmarked and go through every weekday. It usually takes me <15 minutes. Since I started doing this, I’ve been much more up to date and informed on current work. 

First is Twitter. For a long time, I resisted using Twitter because it’s terrible. But after I started following almost exclusively AI safety people and orgs, I’ve found a lot of good papers and ideas. 

Second, I check LessWrong and the AI Alignment Forum :) 

Finally, I look over new titles on arXiv for AIMLComputation and Language, and Computer Vision and Pattern Recognition. More often than not, I find a new paper that I want to click on and check out. 

And as a bonus, I’d of course also recommend these three newsletters

New Comment
7 comments, sorted by Click to highlight new comments since: Today at 1:36 PM

I made this link which redirects to all arxiv pages from the last day on AI, ML, Computation and Language, Computer Vision, and Computers and Society into a single view. Since some papers are listed under multiple areas I prefer to view this so I don't skim over the same paper twice. If you bookmark it's just one click per day!

If you get the daily arXiv email feeds for multiple areas it automatically removes duplicates (i.e. each paper appears exactly once, regardless of cross-listing). The email is not to everyone's taste of course, but this is a nice aspect of it.

I'll use this from now on. 

Are there any Twitter lists you'd recommend for high % of good AI (safety) content? AGI Safety core by JJ (From AI Safety Support)

Feel free to skim through the users I follow at