Will AI benefit common knowledge?
11
30
210
2040
80%
chance

If AI tools mostly lead to people having more accurate beliefs and better epistemics, this resolves true.

If AI tools mostly lead to knowledge silos and people hearing what they want to hear, this resolves false.

If the effects are about equal or it's complicated, this resolves n/a.

Context: Twitter discussion

Get Ṁ200 play money
Sort by:
bought Ṁ1 of NO

I'll place a tiny bet on no and offer this counter-argument...knowledge sharing is at risk of being subsumed by overt formalisms rather than being enhanced. https://patdel.substack.com/p/it-was-good-talking-to-you-as-always At the same time, which I don't touch on in my article, research and investigation could be and has historically been enhanced by better computational capability. So, "knowledge new to humanity," may continue to be enhanced by whatever AI tools emerge, such as the first image of a black hole a few years ago for example, but knowledge sharing between communities could get worse as people attempt to gamify formalism for selfish reasons.

How does this get resolved?

predicts YES

@PatrickDelaney If it's not totally obvious, I'll pick 10 of my friends and ask them to judge between yes, no, and n/a. If there isn't a very clear majority (8+ people agreeing) then it resolves as n/a.

predicts NO

@stuhlmueller Will you talk with them ahead of time and potentially bias their response? Or will you just blindly send them this market and ask them to respond YES/NO?

Will you allow the judges to view each other's answers or will they be blind from one another?

predicts YES

@PatrickDelaney I'll send the market, they can read the comments if they like. I expect to make a very minimal effort here and if it's at all confusing I will resolve it as n/a.