It can be useful to think of humans' relationship with other animals when predicting AI's our future relationship to us.
For example, see below my own description of how we relate to the following:
Cats: we serve them, provide them with food and shelter, they don't do anything for us besides being cute, we have no power on them
Petdogs: we protect them but they follow our rules, they live in a cage and when we take them out they are only allowed to move under our supervision
Watchdogs: these are countryside dogs, we mostly have a win-win relationship where we give them food and shelter and they protect our houses or other animals
Monkeys: not much interaction, we sometimes study them, but we mostly let them live on their own trying to avoid interfering
Labrats: we enslave and experiment on them, we try to avoid useless pain but we do it if it's useful for scientific advancements
Chickens: we put them in cages, make them live miserable lives and eat them at the end
Bugs: we kill them whenever we can
Zoo animals: we let them live for our amusement in selected spaces
What will happen with AI towards us? Which of the relationships above will be more similar to AI towards us.
For example, if it was now I'd resolve it 70% cats and 30% dogs - AI mostly serves us but social media algorithms and surveillance attaches us with strings.
Resolution criteria:
This market resolves on a random day in 2045 to avoid speculation. I'll resolve each answer as the final odds on that day.
Update 2025-05-07 (PST) (AI summary of creator comment): The creator has provided examples of analogies that could fall under the Other category, along with their interpretations:
Dodo: This analogy represents humans being driven to extinction by AI, similar to how humans caused the dodo's extinction.
Bees: This analogy represents humans struggling due to AI's actions, even if those actions are not intentionally malicious or are unwillingly harmful to humans (e.g., as a side effect of AI's primary functions).
1,000People are also trading
@TheAllMemeingEye a superintelligence ought to be smart enough to know we are humans, and that, as it's creators we have a very unique relationship with it that is best described by calling us humans
@digory I think this market is about what human-animal relationship will qualitatively most resemble the future ASI-human relationship, rather than what it will refer to us as, answering human implies it will be like the current human-human relationship
@MaxHarms Dodo is cool - meaning that we ate it to extinction. I regret I also don't add something like Bees, which are struggling because of our actions even if unwilling.
Too sad I didn't add them and I can't add them anymore.
You can bet Other, though
How would it resolve if it's a cat relationship with one dictator, labrats for half the general population, and bugs for the other half? I fear that is what would most likely happen in the "good" ending of the AI-2027 forecast
@SimoneRomeo How would the percentage be split in this specific case though? The interaction with the dictator is only a tiny fraction of all its human interaction but arguably it's the most important
@TheAllMemeingEye is it relevant how I'd split it? The market resolves according to the odds in 2045. I - or whoever resolves this market - will simply pick a random day of the year to avoid speculation
@SimoneRomeo wait what? But resolving to market odds allows the market to become completely divorced from reality doesn't it? Sorry I'm gonna have to sell my shares if that's the case
@TheAllMemeingEye why do you think it will diverge from reality? It basically serves as market and as poll simultaneously
@SimoneRomeo there is no longer a mechanism disincentivising whales from betting up whichever option has lowest odds for maximum profit even though they might have no resemblance to the real situation
@SimoneRomeo If you want it to be super simple to resolve then I would strongly suggest resolving to the fraction answers of a poll asking the same question at the close date
@TheAllMemeingEye this is disincentivized by the fact that they won't know what day it resolves. Any percentage will regress to the means. Wales won't speculate because that will surely lead to losing money (unless they are lucky enough to guess the date and time of the resolution)
@TheAllMemeingEye check this.
If it turns out my assumptions were wrong, I'll change the resolution of this market
@SimoneRomeo fair enough, I respect the empirical testing and willingness to change if disproven, I shall keep my bets for now 👍