Wiki Contributions


[3-hour podcast]: Joseph Carlsmith on longtermism, utopia, the computational power of the brain, meta-ethics, illusionism and meditation

Really enjoyed listening to this. I relate a lot with your perspective on grounding value in our experiences and found Joseph's pushbacks really stimulating. 

DeepMind: Generally capable agents emerge from open-ended play

Is there already a handy way to compare computation costs that went into training? E.g. compared to GPT3, AlphaZero, etc.?

Are there fictional books (novels, poetry, comics) on an EA?

I also really enjoyed the unofficial sequal, Significant Digits.

It's easy to make big plans and ask big questions, but a lot harder to follow them through.  Find out what happens to Harry Potter-Evans-Verres, Hermione, Draco, and everyone else once they grow into their roles as leaders, leave the shelter of Hogwarts, and venture out into a wider world of intrigue, politics, and war.  Not official.

"The best HPMOR continuation fic." -Eliezer Yudkowsky

Books and lecture series relevant to AI governance?

Thanks a lot for compiling this, I'm thinking about switching my career into AI governance and the lists in your Google Doc seem super useful!

Metaculus Questions Suggest Money Will Do More Good in the Future

Cool questions! I am a bit hesitant to update much:

  • they don’t seem to be too active, e.g. few comments, interest count at around 10 (can you see the number of unique forecasters somehow?)
  • the people doing the forecasts are probably EA adjacent and if they did something akin to a formal analysis, they would share it with the EA community, or at least in the comments, as it seems relatively useful to contribute this
On what kinds of Twitter accounts would you be most interested in seeing research?

I’ve seen the claim that economist Alex Tabarrok was significantly ahead of the curve on COVID issues. Would be interesting to see how he did or did not reach and convince the people involved in policy making.

Increasing personal security of at-risk high-impact actors

Just saw this, which seems like a good step in the intended direction:

Canada will become one of the first countries to offer a dedicated, permanent pathway for human rights defenders, and will resettle up to 250 human rights defenders per year, including their family members, through the Government-Assisted Refugees Program.

Building my Scout Mindset: #1

Seconded, I‘d really like to read more of stream of thought inspections like this. Seems like a great practice and also like a cool way to understand other people‘s thought processes around difficult topics.

People working on x-risks: what emotionally motivates you?

Good question. I think I’m maybe quarterway there to be internally/emotionally driven to do what I can to prevent the worst possible AI failures, but re this

Say I'm afraid of internalizing responsibility for working on important, large problems

I always thought it would be a great thing if my emotional drives would line up more with the goals that I deliberately thought through to be likely the most important. It would feel more coherent, it would give me more drive and focus on what matters, and downregulate things like some social motivations that I don’t endorse fully. I suppose one might be worried that it’s overwhelming, but that hasn‘t been a thing for me so far. I wonder if humans mostly deal okay with great responsibilities, which is my spontaneous impression.

(btw, I really enjoyed reading your PhD retrospective, nice to see your name pop up here! I’m doing a PhD in CogSci and could relate to a lot)

Load More