Monthly meeting of EA/ACX/LW community

2 topics for this meeting (we'll spend about an hour on each):

1. Multi-agent cooperation problem: If multiple people need to cooperate to do something (such as running a society), it is beneficial to individuals to "defect" by doing less and letting others do more, or cheat to gain more than their fair share. This may be the root of many of the most dangerous issues in society (including eliminating diseases, corruption, or AI safety). We discuss this with the help of the Moloch lens.

2. Is truth good for mental health? Rationalism states that "that which can be destroyed by the truth, should be", but sometimes false beliefs make us feel good, or keep us from thinking dangerous thoughts - an example of "Bucket Error".