Cold Takes

First Post

Holden Karnofsky -

Welcome to Cold Takes, where it's always fine to read the post later!

About

Most of the posts on this blog are written at least a month before they're posted, sometimes much longer. I try to post things that are worth posting even so, hence the name "Cold Takes."

As of this writing (July 2021), likely initial themes include:

  • Futurism (the next 10,000+ years). What can we say and do today about the very long-run future of the world?
  • Quantitative macrohistory (the last 250-10,000 years). What can we say about very long-run trends in things like quality of life and pace of innovation?
  • Applied epistemology. How should one decide what to believe, especially when (is as usually the case for important topics) one doesn't have time to digest even 1% of the relevant knowledge and arguments?
  • Applied ethics of donations, career choice, etc. What ethical principles should we use to  consider actions like where to donate and what job to take (where there are no clear moral "rules," and instead a question of how to do more good vs. less good)?
  • Links I feel like sharing, often quite old, because they seem important and/or just fun.

A general theme of this blog is what I sometimes call avant-garde effective altruism. Effective altruism (EA) is the idea of doing as much good as possible. If EA were jazz, giving to effective charities working on global health would be Louis Armstrong - acclaimed and respected by all, and where most people start. But people who are really obsessed with jazz also tend to like stuff that (to other people) barely even sounds like music, and lifelong obsessive EAs are into causes and topics that are not the first association you'd have with "doing good." This blog will often be about the latter.

I am the co-CEO of Open Philanthropy and co-founder of GiveWell, but all opinions are my own.

Here are the main series I'm planning so far. As I put things up, I will update the Themes/Highlights page to link to the pieces that have received the most "likes." (Note, though, that I'm the only person with access to the "like" button for this blog.)

  • Most Important Century series. A series of ~10 pieces laying out the case that the 21st century has a high probability of being the "most important" for humanity via the development of transformative AI. This includes discussions of why it looks like we might soon (in the next few decades) develop AI that can dramatically accelerate scientific and technological development, and discussions of why that in turn could lead to a radically different world (in particular, via mind uploading).
  • Has the world improved over time? A series on whether life has gotten better over time and how we should think about the "state of nature."
  • Searching for Atlantis. A series asking mostly, "Is the world doing something wrong, compared to the past, w/r/t innovation?" I emphasize "doing something wrong," not "slowing down" - most discussions online are explicitly about whether innovation is slowing down, but I want to address whether this is about something "going wrong" or just inevitable. "Atlantis" refers to the idea of a past, advanced civilization, now lost.
  • Applied epistemology. Some posts about "how to decide what to believe." These  will include my take on the pros and cons of "explicit Bayesian reasoning" (literally writing down probabilities and values for use in decision making), on how to balance "thinking for oneself" with trusting experts and others, and on how far one should take "self-skepticism" (doubting one's own beliefs).
  • Utilitarian ethics. I'm going to lay out the best case I know for a "hardcore utilitarian" approach to ethics - the basic philosophy-based case for "a large # of small benefits to people can outweigh even a huge harm," "helping enough animals could be better than helping humans," and "reducing risk of extinction, leading to more people getting to exist, could be better than either of those." I'm also going to talk about the weaknesses in this case.