I think this summary is a very good overview of this year's version of the UAPDA: https://www.reddit.com/r/UFOs/comments/1e96n71/the_unidentified_anomalous_phenomena_disclosure/

With the Act having different sponsors, the odds of Lockheed's lobbyists being able to push people to remove provisions like eminent domain as they did last year is low. Biden and all of the majority leaders will likely be lame ducks, and this is going to come up after the election, so people don't have to risk losing the support of the defense industry by doing nothing and allowing it to pass as a rider to the bigger bill it's in.

The hundreds of pages of detail in this bill is immense and clearly how to release this information this will come to take up some of Harris or Trump's time if it passes.

There don't seem to be many people who actually think there should be a reason to keep this stuff secret anymore. Instead, it's more like that nobody ever had a plan for how to get it all out without protests or chaos over previous government actions, and this bill provides a plan.

What qualifies as worldview-shattering? If it turns out that they're extremely capable aircraft decades ahead of what's publicly known, would that count? If it turns out that they are also made by a country not generally recognized as being capable of SOTA aircraft production (i.e. North Korea), not generally recognized as being a country (i.e. Molossia), or previously not generally known (i.e. country of origin has never had a Wikipedia article, Wakanda), or by an Air Force of a powerful non state actor or a conspiracy (i.e. bildeberg group, al qaeda, nazi Antarctic base, SCP foundation), would that count? What if they were of natural or biological origin but required new fundamental physics to explain?

Mostly covered in the linked LessWrong post.

To be extremely specific about the resolution criteria, how would you resolve the market if Eliezer keeps the $150K and RatsWrongAboutUAP's $1K, but you believe (for any reason, however justified) that he should have paid out? I'm only interested in this market if the above scenario resolves YES; otherwise it's partly a market on your beliefs.

This market creator is fair and you have nothing to worry about.

That is not an answer to my specific question.

It’s an answer to a better question. There are always edge cases in markets like these, and Joshua would be the ideal person to handle them.

I find it difficult to imagine any scenario where Eliezer keeps the money and this market resolves no. Maybe if his counterparty dies in an alien invasion?

Agree. Even then he should pay next-of-kin. Perhaps if the worldview-shattering revelation renders money completely pointless?

Thanks, just specifically checking that the resolution criteria will be "NO iff Eliezer pays out" rather than "NO iff Joshua thinks Eliezer should pay out". The second one is much harder to predict, not knowing you.

I'm fine with small probabilities of edge cases (e.g. Eliezer goes bankrupt and says he'd pay out if he could), but I expect a nontrivial probability that the two criteria above will differ because you find something persuasive and Eliezer does not. Can you confirm that you'll resolve this market based on whether the bet is paid out, even if you disagree with his decision?

I think this market should not resolve based on whether the bet is paid out, but rather based on who WINS the bet. If either party backs out from paying, that has no bearing on who actually WON the bet.

@Joshua the funny thing with that tweet is that several of Mira's markets have ended up having questionable resolutions because of under-specified criteria 😬

@PatrickLV I think a more comprehensive summary to account for those edge cases is "Resolves NO iff Eliezer concedes he lost the bet" so we include scenarios where he is bankrupt.

I agree with @benshindel . @Joshua will resolve the market. In fact, I entirely expect that the most likely outcome to this situation is that all of the following are true:

1. "AGI" is achieved before 2029

2. The model is able infer which of the existing data would best convince the 35% who don't agree something is going on and superintelligently persuade the world to actually investigate

3. Eliezer Yudkowsky will be one of the last holdouts, never changing his opinion, just as he never will change his opinion over AI destroying the world

4. Whoever owns the debt will sue Yudkowsky

5. Yudkowsky will declare bankruptcy and be unable to pay this unsecured debt

I never expect any money to ever change hands here, but I do think that in five years the number of people who accept UFOs are world-shattering will rise from the currently polled 65% to 90%.

I do think, though, that @Joshua might benefit from creating a new market on polling thresholds. For example, what percentage of people will agree that UFOs are "real" in 2029? It's already a majority, non-partisan viewpoint, which is not correlated with age or education level. Will it rise to 75%? 85%? That would be an interesting market.

I certainly agree that if that exact sequence of events occurs, this market should resolve NO. As to whether I think that sequence of events is at all plausible… happy to bet on an explicit market.

Polling in what country? Globally? Belief in UFOs is a weirdly US-specific thing, right?

If that entire set is your prediction, can we get a market on it? We've already got some AGI markets, presumably you could use one of those as part of the basis for it.

I'd also be interested in a "Yudkowsky declares bankruptcy (any reason) by $date" market if you wanted to run one.

I would not bet on that market because it is absurdly expensive to sue someone in the United States other than in small claims court, and because I don't have knowledge of Yudkowsky's assets.

Would I sue someone for $150,000? Absolutely not. When I wanted to sue BlockFi CEO Zac Prince for $2.68 million, I was quoted $800,000 in legal fees. Additionally, I have never found a single lawyer who is more competent than I am at the law, and I'm not a lawyer. There's actually been times when lawyers have been asking me to explain the theory of the case (before I stopped using them.) And the legal system is rigged so that in many cases going pro-se is prohibited.

Lawyers want to be paid hourly and prior to commencing any initial work, which means their incentive is to draw out the case and not do a good job. Even if they did do a good job, it doesn't matter how much you are owed if you can't afford to pay up front. The bar for filing a profitable lawsuit is somewhere around $10,000,000. That's why, far below, I was willing to bet big on this but would only accept a first lien against a house as collateral.

Does Yudkowsky really seem to you like the sort of person who would ever acknowledge he was wrong? Whoever owns this unsecured debt probably has no written contract associated with the bet. There's no way he will ever pay a dime.

Doesn't this whole chain of events seem unlikely even in the scenario where the UFOs are real? Wouldn't most of the cases where this market resolves NO be because the UFOs do something? Or are the UFOs incapable of ever having any meaningful effect other than AGI deciding that they are real and to resolve the bet?

It's just got an invisible garage dragon feel to me

Does Yudkowsky really seem to you like the sort of person who would ever acknowledge he was wrong?

Yes, and if you write a "will Yudkowsky admit he's wrong about anything at all by 2029" market that doesn't sound too weasel-wordy, I'll bet on that too. (Ambiguous stuff that might scare me off betting would be anything with subjective judgment about whether the admission is about an "important" thing.)

The bankruptcy example was just an example of an edge case, not something I consider to be more than 1% probable. The crucial question for me was how Joshua would resolve the market if Eliezer does not concede, but Joshua thinks Eliezer should concede (because I expect that scenario to be at least 10% likely). Now that Joshua has confirmed he would resolve the market YES in that case rather than NO or N/A, I feel better calibrated on the odds.

I don't expect UFOs to ever do anything. The most likely explanation for UFOs is that some portion of all unexplained weird stuff people have seen throughout history has been caused by non-human intelligence, not just "aliens" recently.

Look at Stephen Wolfram's theory of computational reality. We compute things that we view as interesting, and these non-human intelligences have always been computing things they find interesting, and those two interests have no intersection whatsoever. We are so different from them that the concept of "communication" might not even make sense to them.

It's not like they are "a million years more advanced than us." They might not even have a concept of time. They're not coming down out of the sky to make an announcement or take over the planet, if that's what people are expecting.

If Yudkowsky dies before 2029 then he got the $1k before he died, so he has won the bet, I think?

It would be hilarious if his estate couldn't be closed because of possible UFOs but I don't think that's how it works.

You can participate in an academic research survey on your thoughts on extraterrestrial contact:

https://www.durham.ac.uk/departments/academic/law/news-and-events/news/2024/july/academics-at-durham-law-school-and-st-andrews-university-launch-global-online-survey-/survey-invitation-english/