So don't take the lack of publically disclosed evidence as an indication that no evidence exists, because it isn't.

It actually is, in the sense we use the term here.

Surely that depends on exactly what you define "friendly" to mean.

It certainly seems to. Somewhere on my list of "ways to stop an AI from torturing me for 10 million years" is "find anyone who is in the process of creating an AI that will torture me and kill them". I'm not overly concerned what name they give it.

The would be more likely to work if you completely took out the 'for existential risk' part. Find a way to cooperate with people effectively "to make money". No need to get religion all muddled up in it.

I don't think Bayesianism is a recipe for action in the first place - so how can "pure Bayesianism" be telling agents how they should be spending their time?

It tells them everything. That includes inferences right down to their own cognitive hardware and implications thereof. Given that the very meaning of 'should' can be reduced down to cognitions of the speaker Bayesian reasoning is applicable.

If I commit quantum suicide 10 times and live, does my estimate of MWI being true change? It seems like it should, but on the other hand it doesn't for an external observer with exactly the same data...

It makes no difference. You're either throwing away Everett branches or having a chance of throwing away everything. This experiment doesn't tell you which. You could, however, conclude that you're a damn fool. ;)

Do you have a reason of sarcasm?

It felt like irony from my end - a satire of human behaviour.

As a general tendency of humanity we seem to be more inclined to be abhored by beliefs that are similar to what we consider the norm but just slightly different. It is the rebels within the tribe that are the biggest threat, not the tribe that lives 20 kms away.

I hope someone can give you an adequate answer to your question. The very short one is that empirical evidence is usually going to be the most heavily weighted 'bayesian' (rational) evidence. However everything else is still evidence, even though it is far weaker.

But now I sense an even more disturbing definition: rational as opposed to empirical. As I use scientific evidence as the most important arbiter of what I believe, I would find the anti-empirical idea of 'rational' a big mistake.

Indeed. It is heretic in the extreme! Burn them!

I seem to have a somewhat more cynical outlook. Judging real humans by the criteria of the sorting hat would result in far more Slytherins than members for the other houses.

If that were so it would defeat the whole point of placing anyone in Slytherin to become one.

Yes, more or less. Unless there is some reason you want people to become better Slytherin.

You could - "with a lot of additional adjustments". You would to have to actually work at turning them into Slytherins, and doubly so if there are no natural Slytherins there at all to lead the way. And probably not everyone anyway.

My claim is that most humans outside of fairy tales already are Slytherins.