Regularly meta-optimization

post by Crazy philosopher (commissar Yarrick) · 2024-06-25T06:12:58.939Z · LW · GW · 6 comments

Contents

6 comments

The usefulness of the different actions differ by orders of magnitude. Sometimes, redirecting efforts can increase your efficiency by an order of magnitude or more.

Imagine that the person who first came up with the idea of sorting charitable foundations by efficiency, instead of implementing it, went to wash the dishes. Or Eliezer decided that creating a community is long and strange.

I'm not sure about the others, but I'm discarding many ideas, including potentially extremely effective ones, because I'm using the absurdity heuristic (this idea will change the world too much to be true). Or I reject ideas, because in order to implement them I will have to leave my comfort zone, and it causes me negative emotions.

Therefore, I came up with a technique that I call Regularly Meta-Optimization. It consists in regular trying to find potentially extremely effective ideas among those that you have been thinking about lately.

One of the potentially very effective ideas for me is to share a few of my potentially very effective blog ideas that can be implemented by anyone.

1.  Creating a rationalistic YouTube channel. If we convince at least 20% of people of the danger of uncoordinated GAI, we will probably be able to get the government regulation we need in this area. The spread of rationality can also lead to very large positive side effects, for example, by radically increasing funding for anti-aging research and effective charity funds. This is feasible because rationalist ideas are most likely interesting to a significant number of people when presented correctly, because these are our ideas that are true, and therefore they are more convincing.

6 comments

Comments sorted by top scores.