Will Eliezer Yudkowsky win his $150,000 - $1,000 bet about UFOs not having a worldview-shattering origin?
828
10kṀ6.9m
2028
92%
chance

Original Lesswrong thread here.

Original tweet here:

Unlinked market with shorter timeframes here: /Joshua/when-will-we-know-that-any-past-ufo

Get
Ṁ1,000
to start trading!
Sort by:

I have been betting this market up to around 7-8%.  I, along with others betting alongside me, often get criticized and dismissed for being 'overconfident' about the existence of a conspiracy.  Does anyone else recognize the irony in that?

To be clear, I am expressing a 7% confidence that perhaps there are people in the world that know quite a bit more than me about topics which would be ontologically shocking to me.

It appears that the side betting against me is at least 92% confident that they have all the information necessary to know that there are no groups that have information that would be ontologically shocking to them.

Can someone help me understand the technical justification for such a high confidence?  It appears it's largely based on social conformity (appeal to authority).  To my knowledge, there is no proof that exists that this is not the case.  Since there exists no such proof, a rational person should assign some > 0 probability to the possibility that they might be wrong.

Are there any 'no' bettors that are willing to share their actual confidence that they assign to themselves being wrong?

@Krantz
- World shattering stuffs are rare.
- If you condition on UFOs implying something world shattering, you get, I think, a very small fraction of the set of things that could be world shattering.
- If you condition on it being somewhat successfully hidden to the public, it is still a smaller set (because there need to be a good reason to hide it, and the capacity).

I think there are less than 1% chance, for all these reasons and some others.

But I agree I am the one being confident here, not you. So yes I see the irony of reproaching your overconfidence.

@Krantz to add to dionisos's response:

I think a 7% probability is somewhat reasonable for "there are people in the world that know quite a bit more than me about topics which would be ontologically shocking to me."

But that's not what you're betting on in this market. Here, you're betting on "there are people in the world that know things about UFOs specifically which would be ontologically shocking to me, who have successfully kept those things hidden thus far, and those things will become public knowledge within the next three years (OR: there are no such people, but humanity as a whole will learn new ontologically shocking things about UFOs in the next three years)".

Tossing out some off-the-cuff probability estimates (each conditional on the previous ones):

  • "UFOs have an ontologically shocking origin" (1%)

  • "people have strong evidence of UFOs' ontologically shocking origin" (10%)

  • "those people have been able to contain that information" (10%)

  • "those people will not be able to contain that information for three more years" (2%)

This leads to an extremely low probability for this market resolving yes as a result of a conspiracy being exposed.

Maybe you think that the next three years specifically are an unusually likely time for any such hidden knowledge to become public (e.g. because of AGI/ASI). But even turning the "will not be able to contain that information" probability to 100% gives a quite low estimate for this market resolving yes.

I'm curious what probabilities you'd give for the statements above.

@jcb also of note:

If an alien IFO arrived tomorrow, this market still resolves NO unless they prove that they were responsible for one in the past.

@jcb @dionisos

1. World shattering stuffs are rare.

I would agree, but not because there is a lack of worldview shattering discoveries to be made. It's because of a human bias to focus on topics within one's own worldview. As such, I don't see a reason to reduce my expectancy to encounter worldview shattering discoveries. I only see a reason to expect that nobody else will believe me when I encounter them. That seems to track with all the numerous sightings nobody believes.

2. If you condition on UFOs implying something world shattering, you get, I think, a very small fraction of the set of things that could be world shattering.

I'm not sure I understand this claim. Are you just saying that UFOs being known to exist without a prosaic explanation is just one of many worldshattering truths that could be known? I don't see how that would compel me to reduce the likelihood that UFOs have a non-prosaic explanation. I do see it as a good reason to believe there are probably even bigger world shattering updates beyond UFOs that I can't even imagine.

3. If you condition on it being somewhat successfully hidden to the public, it is still a smaller set (because there need to be a good reason to hide it, and the capacity).

It doesn't seem like it's being "hidden" very well. I believe the noise that exists in society's communication infrastructure coupled with the ostracization of the topic simply presents the information from either being found or believed when found (at scale). All that's really necessary to maintain the status quo worldview is for the government not to confirm it and let human nature maintain the status quo.

4. UFOs have an ontologically shocking explanation.

This either seems like it's tautologically true or it is missing the crux of the market. For example, if the Whitehouse gave a press conference tomorrow and truthfully said, "UFOs are real, we've retrieved many crashed UFOs and have been secretly studying them in underground bunkers for decades trying to replicate them but still don't have an explanation for what they are, how they work, or where they came from.", that would (1) seem to be an ontologically shocking update to the status quo worldview of most people and (2) would compel a 'no' resolution to this market.

In other words, 'no prosaic explanation' seems to be an 'ontologically shocking explanation'. The crux of the market seems to be 'has the government (or equivalent) been secretly studying UFOs that have no prosaic explanation?' (and will they officially tell us that in the next 3 years?).

5. People have strong evidence of UFOs' ontologically shocking origin.

This seems to have the same problem as 4. Perhaps a charitable paraphrase would be 'People have strong evidence that UFOs are real and have no prosaic explanation.'?

6. Those people have been able to contain that information.

Same as 3 above.

It doesn't seem like it's being "hidden" very well. I believe the noise that exists in society's communication infrastructure coupled with the ostracization of the topic simply presents the information from either being found or believed when found (at scale). All that's really necessary to maintain the status quo worldview is for the government not to confirm it and let human nature maintain the status quo.

7. Those people will not be able to contain that information for three more years.

A charitable paraphrase might be, 'The general public will not be open minded enough to sort through the noise produced in the public discourse'.

Out of all the reasons both of you have provided, the only one that feels like justification for increasing my expectancy of a 'yes' resolution is my paraphrase of 7. I think 'the general public not being able to sort through the noise produced in the public discourse' is highly likely. It's also probably why nobody is taking existential risk from AI seriously either..

I'd really like to fix that.

Anyway, thanks for the response.

bought Ṁ100 YES

@Krantz The technical justification is just that this:

It appears that the side betting against me is at least 92% confident that they have all the information necessary to know that there are no groups that have information that would be ontologically shocking to them.

Is straightforwardly false. I think that out of hundreds or thousands of possible cover-up conspiracies of roughly this size, at least one is probably true, but each individual one is almost certainly false. This conspiracy fails for the same reason as all other conspiracies: too many people would have to keep too big a secret. I'd put the chance that there's an extent coverup at maybe 0.05%, and the chance that later analysis will reveal something not yet known to anyone at maybe 0.2%.

@Ansel I didn’t read this whole thing but this shows a lack of basic knowledge.

@JasonQ Yeah agree he's conflating two different concepts there. Doesn't mean the whole report is wrong, or mean anything at all about the report itself in fact. The report should be judged on its own merits, not on the analytical skills of a guy on the internet who got excited about it.

@Ansel my base assumption is that if a more careful analysis from people who know what they are talking about supported the wild assertions he jumps to, you’d post that rather than some confused guy. That does tell me information about the report.

filled a Ṁ50 YES at 99.0% order

Easy YES. Pete Hegseth would have yapped to his wife about it already.

bought Ṁ50 YES

Does anyone know how are loans calculated today? I bought 10k, but my daily loan did not increased. Are they now capped? Per market or globally?

I am asking here specifically, because loans have big impact on price here.

(108140)

@Irigi the loan system is very generous (with minimal "risk management"/oversight), but the maximum loan you can receive for any single market is 5% of your net worth. Your position this market is 117% of your net worth, & your current loan on this market is 88% of your net worth, so that means you can't receive further loans on this market (but you will still receive loans normally on all the other markets on the site—loans aren't intended to endlessly increase leverage for people betting on a single market).

(the per-market 5% of nw cap should be added to the UI)

there's also a cap on the total leverage per account that blocks new loans entirely, but it's extremely high and i don't think you're close to reaching it.

@Ziddletwix how do you know how big Irigi's current loan on this market is? Loans have always seemed quite opaque to me and I'd be keen to monitor my own loans more closely and keep track of which markets I can get more leverage on and which I can't

@Ziddletwix Thank you for the explanation. I was not complaining, I just want to update my knowledge on how the system works now and act accordingly. I think the cap per market is a good thing.

@Ziddletwix Btw. might be a nice feature to show user on what market they reached the loan cap. (Usually, they want to take that into account).

@Irigi agreed! It’s under explained and that’d be a good feature

@Fion you can see the loan info for each market if you click to see more under the trades tab of their profile (the info is probably generally available in the API but that’s how you check in the UI)

@Ziddletwix wait... how is it under the trades tab? I don't see that column option

@bens it’s not a column. You need to click to expand the trade itself

Do unidentified "flying" interstellar objects count for the purposes of this market?
https://thebsdetector.substack.com/p/one-aliens-trash-is-another-aliens

@bens

I think we’d know any candidates for that gray area by now.

This is so funny

@metacontrarian @JonasVollmer if you're looking for a free money volatility printer, I can recommend a better market, but after reading AI-2027 I think you might not like it XD

https://manifold.markets/ScottAlexander/in-2028-will-an-ai-be-able-to-gener?r=YmVucw

© Manifold Markets, Inc.Terms + Mana-only TermsPrivacyRules