Extraordinary ethics require extraordinary arguments

Butterfly effects essentially unpredictable, given your partial knowledge of the world. Sure, you doing homework could cause a tornado in Texas, but it's equally likely to prevent that. To actually predict which, you would have to calculate the movement of every gust of air around the world. Otherwise your shuffling an already well shuffled pack of cards. Bear in mind that you have no reason to distinguish the particular action of "doing homework" from a vast set of other actions. If you really did know what actions would stop the Texas tornado, they might well look like random thrashing.

What you can calculate is the reliable effects of doing your homework. So, given bounded rationality, you are probably best to base your decisions on those. The fact that this only involves homework might suggest that you have an internal conflict between a part of yourself that thinks about careers, and a short term procrastinator.

Most people who aren't particularly ethical still do more good than harm. (If everyone looks out for themselves, everyone has someone to look out for them. The law stops most of the bad mutual defections in prisoners dilemmas) Evil genius trying to trick you into doing harm are much rarer than moderately competent nice people trying to get your help to do good.

Personally, I just dismiss scrupulosity as an error. I don't need a justification for doing this, any more than I need a justification for concluding that if, when doing some mathematics, I derive a contradiction, then I must have made an error somewhere. Call this the absurdity heuristic, call it a strong prior, but obsessing over the unknowable potentially enormous consequences of every breath I take is an obvious failure mode, and I don't do obvious failure modes. Instead, I do what looks like the right thing for me to do, the thing that only I will do. (That is just a rough description of a rule of thumb, not something with a detailed philosophical analysis behind it.)

This probably makes me a bad person by the standards of the typical EA. I have no interest in dissuading them from their calling, but it is not my calling.

A way I would suggest looking at it: your scrupulousity daæmon has its own estimates of prior probabilities for such things as “you being a fundamentally decent person” and “you being on the cusp of accomplishing dreadful evil, maybe by accident”, and those estimates are, respectively, low, and high, because the dæmon shares your mental substrate and inherits the effects of depression.

As with us, the dæmon’s priors guide its theorising. It expects you to accomplish dreadful evil. How? Well, if your prior probability for something is high, and there isn’t a simple explanation, then there must be a complicated explanation! And people are very good at producing complicated hypotheses out of nothing, we’ve been doing it for millennia.

One thing that's helped me is to more proactively cache these internal dialogs, and when they start again to ask myself "do I have any new significant information on this topic compared to last time I spent a few hours on it"? If no, I can skip to the end and use the one-time pass before I'm exhausted.

Also, after enough repetition and variation, I'm pretty convinced that "it all adds up to normal" applies to non-extreme ethical choices as well as it does non-extreme physics.

To generalize this heuristic a bit, and to really highlight where its weaknesses lie: An ethical argument that you should make a significant change to your lifestyle should be backed up more strongly in proportion to that change.

For example, to most people, the GWWC 10% pledge is a mildly extraordinary claim. (My parents actually yelled at me when I donated my Christmas money at 17.) But I think it does meet our bar of evidence: 10% income is usually no great hardship if you plan for it, and the arguments that the various EAs put forward for it are often quite strong.

Where this heuristic breaks down is an exercise to the reader. :)

A few ideas:

A. Utility maximization - if you donate a bunch of money to an effective charity and they save 100 lives, but in the course of delivering say, medical supplies to somewhere (which is how 100 lives were saved) one of the volunteers falls out of a boat and drowns:

1) It couldn't have been predicted in advance.

2) The gains out weigh the benefits.

So, the real question shouldn't be "Will something bad happen?" But "is it remotely plausible that the bad will outweigh the good?" Yes doing homework might, once in a million person years, result in some terrible tragedy, and if you're really worried about paper then go electronic. The scenario I outlined sounds kind of ridiculous. Most actions don't have big impacts by themselves/destroy the world (or have tremendous impact). If you determine the course of action with the highest utility, it might not be as high as you think, but it's still the best thing you can do. (Using time effectively might involve lots of planning, but that might be less important (currently) than doing what you know you have to do right now.)

B. Is the "Skeptic Demon" helping? 'Cause if it's never done anything useful, and if it's getting on something it's got on before, it probably doesn't have anything new to say. (Writing stuff down might help. It seems like it'd be more useful to apply to inaction, if that makes sense?)