@an1lam I personally plan to live forever just so I can see my Metaculus predictions resolve.
It's speculative, but I would not rule out that Musk may be more optimistic than 60%, even if publicly he says 60%. There is a bit of an incentive to give lower probability than actually expected, I think: if he says, say, 80% and fails, that would probably be seen as a bigger failure than 60%, whereas success will be seen as a lesser one.
Incentives are different with long-term predictions, I think.
@AngraMainyu I think the argument is roughly "Breyer is going to resign in 2021 or 2022 while the Democrats control the senate and the Presidency".
Recall campaign claims to have gathered 1,825,000 signatures as of this morning.
Based on an 82.1% validity rate, that would be sufficient for the recall to happen.
— edited by cd
@Jgalt Whether it enables FTL travel or not would have a huge effect on the probability, so I think we should specify this. IMO, an Alcubierre drive that is slower than slight is not particularly interesting — the whole point is to allow FTL travel.
So I'd suggest we make the criterion "Alcubierre-like" and FTL.
I don't understand 2300 as the resolution date. If in fact there is life on Mars, the overwhelming likelihood is that it would be found fairly quickly. To make 2300 as the resolution date seems really unfair to those who think the answer is no.
Therefore please consider making the question, "Will we find life on Mars by 2030?"
Are we excluding the pattern where every sentence is cherry-picked by a human editor? E.g. generate 100 candidates for sentence 1, select one, then generate 100 candidates for sentence 2, select one etc. In this case, no sentence needs to be edited by a human, just selected from among entirely AI-generated candidates. This is the most likely pattern an “AI-written” book would actually take IMO.
I'm noticing that there's more predictions on this question than on this related question asking when Starship will land properly. If you haven't had a chance to predict on that one yet, get your predictions in!
@Pablo I'm so happy to see this question again! And now in hue!
Given that Ultimate Gray was also in the New York / London overlap for 2021, but I missed the relevance of the 'core classics', I'll be predicting on the basis of the total overlap of Spring/Summer reports this year. Unfortunately, this approach means I won't be able to narrow my prediction until September. I don't view the Autumn/Winter trend reports (which were released a few weeks ago) as relevant to this forecast.
There's something fairly nasty about predicting maximums on Metaculus. The "correct" distribution is a mixture of discrete (probability we've already had the maximum) and continuous (probability the maximum is yet to come). (The distribution also tends to be very not-symmetric, which is always quite hard to predict using the Metaculus input).