but if, noting that there are no such bullet points, you make the decision not to read the body text -- you should probably refrain from commenting on the body text. A license to opt-out is not the same thing as a license to offer serious criticism. Eliezer may be wrong, but he's not stupid, and he's not crazy.

This is sound general advice, but at least one observation makes this situation exceptional: Yudkowsky's conclusions about ethics are never summarized in terms that contradict my take. I don't think your rendition, for example, contradicts mine. I'm certainly not surprised to hear his position described the way you describe it:

Anyway, saying that morality is a tool doesn't get you out of the original need to decide which lifestyle you want in the first place. Should I be an artist, or a politician, or a banker? Why? Eliezer's answer is that there are no shortcuts and no frills here; you check and see what your brain says about what you 'should' do, and that's all there is to it.

Now, I don't think the decision of whether to be an artist, politician, or banker is a moral decision. It isn't one you make primarily because of what's ethically right or wrong. To the extent you do (and in the restricted sense that you do), your prior moral habits are your only guide.

But we're looking at whether Yudkowsky's position is intellectually respectable, not whether objective morality--which he's committed to but I deny--exists. To say we look at what our brain says when we're fully informed says essentially that we seek a reflective equilibrium in solving moral problems. So far so good. But it goes further in saying brains compute some specific function that determines generally when individuals reach that equilibrium. Leaving aside that this is implausible speculation, requiring that the terms of moral judgments be hardwired--and hardwired identically for each individual--it also simply fails to answer Moore's open question, although Yudkowsky claims he has that answer. There's nothing prima facie compelling ethically about what our brains happen to tell us is moral; no reason we should necessarily follow our brains' hardwiring. I could consistently choose to consider my brain's hardwired moralisms maladaptive or even despicable holdovers from the evolutionary past that I choose to override as much as I can.

Robin Hanson actually asked the right question. If what the brain computes is moral, what does it correspond to that makes it moral? Unless you think the brain is computing a fact about the world, you can't coherently regard its computation as "accurate." But if not, what makes it special and not just a reflex?

I do feel a bit guilty about criticizing Yudkowsky without reading all of him. But he seems to express his ideas at excessive and obfuscating length, and if there were more to them, I feel somewhat confident I'd come across his answers. It isn't as though I haven't skimmed many of these essays. And his answers would certainly deserve some reflection in his summation essay.

There's no question Yudkowsky is no idiot. But he has some ideas that I think are stupid--like his "metaethics"--and he expresses them in a somewhat "crazy" manner, exuding grandiose self-confidence. Being surrounded and discussing mostly with people who agree with him is probably part of the cause.