This is why I consume so many books and articles even though I don’t remember most of their specific content when asked about them. I’m also usually doing a breadth-first search to find candidates that might be worth a deep dive. E.g. I think I found Ian Morris faster than I otherwise would have because I’ve been doing breadth-first search.
Notable lessons (so far) from the Open Philanthropy Project.
Tegmark, Russell, and Horvitz on the future of AI on Science Friday.
Eric Drexler has a new FHI technical report on superintelligence safety.
Brookings Institute blog post about AI safety, regulation, and superintelligence.