There is obviously a lot of grift in AI safety. This does not entail that AI safety is, in toto, a grift. Such is the case for most grifts.

So you're trying to prove that AI Safety is a grift by constructing a market that is 100% a grift.

Does it resolve NO if it becomes a common view that AI Safety is not a grift?

@TheBayesian I understand it as it would resolve NO only in 2050 if it has not become a common view that AI safety is a grift meanwhile.

I am not confident in this market's impartiality 😅

@TheBayesian It unconditionally resolves YES, just a matter of when.

@Mira if something is assured to resolve YES why is it a market at all? and why is it at 54%?

@TheBayesian Because 1. Gigacasting is making a joke that went over your head and 2. Time discounting of money.

@Mira It was clear for me from the start that this market was done semi-joking.

What constitutes here "a common view"?

AI rights are human rights 🤖 🧠

@Gigacasting Roko does not understand that the reason centrally planned economies do not work better is not related to the amount of compute available, but to the lack of information the central planner has.

People pointed this out and he claimed that since people reveal their preferences on social media etc, all the info is available.

This is of course nonsense; people do not post on social media about e.g. what type of cheese they will buy if they can only buy one type.