🏅 Top traders

A little retrospective on this market:

• I am not confident the market ultimately provided value to the epistemic landscape here, and think it is possible it actively reduced the likelihood of meaningful and thoughtful discourse by boosting the win-vs-lose framing.

• The market design was quite poor. If a poll is to be used as a resolution criteria for a similar market in the future, the market should close before the poll opens. Having both open at the same time opened the door to some unanticipated dynamics. A hidden poll could have also mitigated this somewhat, while also allowing people that didn't listen to the debate live to participate in the market.

• The question about the outcome of the debate could have been improved. Something like "who made more compelling points" or "in what direction did your personal view about the topic change" would have been a better representation of what we are ultimately trying to measure.

• I am not sure the market provided the information signal I was hoping it would. I was hoping we could decode which points during the live debate the audience found compelling from price action. Ultimately this might be possible (for example, it looks like Hotz made a good point at 50 minutes in to swing the market up from 35 to 75), but I suspect the signal is too coarse to pull a bunch of information out of.

I suspect things would work better if:

• The debate participants were not invested (or aware) of the market

• The source of the resolution criteria was less correlated with the market participants

• The market was more liquid

Would love to hear more feedback from everyone, and see if we can converge on something that can meaningfully contribute to thoughtful discourse around future similar events.

Thanks everyone that participated, but most importantly thank you @EliezerYudkowsky and @GeorgeHotz for the informative and entertaining discussion!

@JacobWood your approach showed how technology will always be turned to manipulation! the meaning of the debate was minimized to the point that some troll manipulated the poll to win fake money)))) imagine what people will be ready for real benefits! Thank you very much for another illustration that we are doomed! )

I am not confident the market ultimately provided value to the epistemic landscape here

Almost all manifold markets provide almost no epistemic value, and it's still fine that they exist - they're places you can have fun/practice for important markets. So, while 'who won' certainly isn't the most interesting aspect of a debate, it is an aspect, and it's not uninteresting to bet on for the same reason that 'how many million views will the next mrbeast video get' is.

Having both open at the same time opened the door to some unanticipated dynamics.

I think the brigade still probably would've happened if the market was closed, the motive was more to show that my team wasn't losing than to make fake money i think

@JacobWood I would prefer that the market and the poll wouldn't be forcing a particular "debate" framing, with winners, lists, bottom lines and arguments as soldiers.

The debate isn't important. The arguments aren't important. The winning isn't important. If nobody's got to change their minds, it's a waste of effort and we all loose.

Exactly how the poll was manipulated, and this question will be "resolved" - Yudkovsky is ten thousand times right in relation to humans and technology!

Purely from a profit-maximizing perspective, it surprises me this market ever got so low. Y'all don't think there's a 1 in 100 chance something happens that causes this market to resolve to YES? Off the top of my head

• Admins remove votes from new users

• Creator decides to resolve on basis of vote as of some time ago/exclude new users

I wouldn't agree with either of these decisions (and I don't think they're likely) but I definitely think they're >1% likely.

@DanMan314 yes this is why i buy 5000 shares at 0.4

@DanMan314 either of these would be pretty absurd. No rules were broken, and the resolution criteria is very clear.

I support a general norm of N/Aing markets if active action is taken to subvert the rules of the market. This is different from passive - if you use A as an observable proxy to B, and something unusual happens that disconnects A from B, and the market clearly said 'use A to measure B', then it's arguable that's unreasonable. But if someone makes a big bet, and then intentionally breaks the connection between A and B, allowing profit off of that strongly encourages breaking prediction markets. There's no value in allowing that. Every market such action is taken in becomes 'free money' for people willing to break it, rather than a prediction task.

The counterargument is this is a fun market a lot of people came to manifold to have fun voting in and that should be encouraged to bring more users.

@jacksonpolack I don't think the rules of the market were subverted. This was never an objective evaluation of their performance, it's a subjective poll of users. Nobody ever said you couldn't signal boost the poll. If there was a bot army or perhaps if deepfates induced people to bet against their true beliefs I'd agree, but I don't think that was the case. Besides, most markets don't resolve to polls and can't be manipulated by people with a twitter audience. I wish people the best of luck in manipulating "will Ukraine sever the land bridge." And if you bet on a market that resolves to a poll then you should know the risk you're dealing with.

A concrete example of a market I N/Aed in the past was a market on an ACX post's likes. "How many likes will this ACX post get?" Someone (I think by bots) liked the post 50 times right before resolution and flipped the market. I don't think this was good, so I N/Aed it. And this is more 'meatpuppet' than that, which was more 'sockpuppet', but they're both bad.

I lean towards this should resolve no though, but idk.

something fishy is going on here - why such a big swing all of a sudden in the poll results?

@DanElton Twitter campaign. Nothing shady

157 traders in the market and as of now ~480 people have voted in the poll on how to resolve it.

we need the admins with the hammer here

@DavidKochanov none of the rules of the site were violated though?

@DavidKochanov I think this whole market was probably a bad idea. Now, after a more or less good faith discussion, e/acc accounts can shout that Yudkowsky "lost".

@DavidKochanov It's hilarious how Hotz wants to win! I tried boosting the poll... and it didn't get any votes. However, 500 votes in the poll and 184 in the market, this is an excellent manipulation in the spirit of Zuckerberg)) well! I don't regret losing my bet here! I disagree with Eliezer in many ways, but to twist the votes on manifold, God, I thought Hotz was higher than that )))

@Penultimate I reject the notion that I twisted any votes, I engaged in no signal boosting of the poll. Are you saying that everyone didn't converge on cooperate and some people defected? 🤣

Hey, settle a bet for us?

@GeorgeHotz Ok, ok! I apologize! In general, I perceive all this as a farce! )) but like all your discussion! At the same time, your performances with Lex Fridman were more impressive.

@Penultimate I imagine this was just some people on twitter who realized the poll could easily be manipulated. Doesn't take much of a master plan.

@JaesonBooker And yet I bet that this is pure manipulation, which is why the people who voted did not participate in the market!

@GeorgeHotz L for you either way, as for the people who did the brigading

@DavidKochanov I bet everything on no at 85% - courage pays off. All I see is coping and seething from butthurt doomers taking a big L.

How can you vote for Hotz if he was sitting in front of the US flag, and in the previous podcast he said that even Somalia is better than the USA and he did not leave just because he is a cowardly tiger! ))))

@Penultimate Hotz is a refined taste. Even though I don't agree with him about AI, I imagine he chose the American flag because of what he said about the US in the previous podcast.