What if it’s literally the “biggest” guest so far?

@mqudsi

@PeterF How would we know it’s not him (latest episode)?

What the hell is going on with this one? @Mira we know you like to pull pranks but maybe you actually have some insider information. In what world are they the biggest guest though?

PS: I do understand that they/he is admired in the community, including by Dwarkesh, but I'm not sure I understand how that justifies such a difference in pricing with many other options below.

@NicoDelon i happen to have very insider info on why people think this is likely, and even more insider info on why it might still not happen

@MattF so intriguing!

@MattF hm I don’t doubt that gwern could be a guest, but I have extreme doubt that he would be the one referred to in this market

@mods the new Jeff Bezos is duplicated, rename to invalid

done

Surely it isn't referring to Daniel Yergin, whom I know basically nothing about? I'm wary of betting against it without digging in, but that feels facially wrong, no?

@HarrisonDurland Yeah, just buy NO.

Market should resolve based on the BMI of the next guest. Period.

There's two Bezos now, Bezos and Jeff Bezos

@DylanBowman How would you handle the following situation?

• The guest is not listed, so Other would resolve YES

• The podcast is released and the guest’s identity is revealed

• A trader adds the guest as an option, which means Other now resolves NO

• The trader then profits by buying a shitton of NO

@biased I guess the timestamp of the newly added option should be checked? If the reveal of the guest's identity was before the added option, then the added option should be N/Aed.

@biased This should be handled by Manifold.

That doesn’t address my proposed situation, which said that a fast-fingered trader would add the real guest as an option then short Other into the ground and make a ton of profit.

The Other option is easily exploited when a market doesn’t have a known resolution time, as an option can always be added for the revealed true thing once the event has happened.

@biased Ah right. Misunderstood.

@biased All of those who bought Other will get shares in the newly added option, so they wont mind. The one who adds the option will get a mana boost by adding a new option compared to just betting up Other, but most of the mana will be made by betting the correct option (either specific person or Other) to 100% anyway as in any other market. So I wouldnt lose sleep over it.

@biased There is no problem with the other option. The only thing is that once the guest is revealed, someone can take all of the liquidity pool

@biased I think you're misunderstanding how Other shares work. If I buy 500 shares of "Other" and then someone adds, let's say, Blorg Shmorgus as soon as they realize Blorg Shmorgus is the "biggest guest", then I still get 500 shares when that resolves YES.

There's no difference between a fast-fingered person reacting to the news, creating Blorg Shmorgus, and betting everything else down than a fast-fingered person reacting to the news that it is, say, Barack Obama, and betting that up and everything else down.

So sure, the first person to react to news profits! But this is not unique to new answers being split off from Other.

One thing I find remarkable is how few women he’s interviewed. I counted five total since the beginning of the podcast. I don’t know what that says about the answer to this question but it’s noteworthy in itself.

@NicoDelon He talks to world class:

AI researchers

Historians

Economists

Financiers

Politicians

I wonder if 5/72 (7%) is a in line with what the distribution of *world class* people in those categories is. If women are not better represented than that ratio in the top end of these fields, then Dwarkesh shouldn't be interviewing them at a higher ratio than that.

At NeurIPS (most prestigious AI conf), only 3% of award winning authors are women

Only 3% of Nobel Prize Laureates in Economics are women
https://en.wikipedia.org/wiki/List_of_female_Nobel_laureates

A study examining popular history books published for general readers in 2015 found that:

• 75.8% of the total titles had male authors, meaning approximately 24.2% had female authors.

@MattF Interesting subtle move from statistics to what he ‘shouldn’t’ do.