←back to thread

AI 2027

(ai-2027.com)
949 points Tenoke | 4 comments | 03 Apr 25 16:13 UTC | HN request time: 0.704s | source
Show context
Jun8 ◴[03 Apr 25 20:41 UTC] No.43575119[source]▶
ACT post where Scott Alexander provides some additional info: https://www.astralcodexten.com/p/introducing-ai-2027.

Manifold currently predicts 30%: https://manifold.markets/IsaacKing/ai-2027-reports-predictio...

replies(4): >>43575216 #>>43577447 #>>43577767 #>>43582073 #
1. crazystar ◴[03 Apr 25 20:51 UTC] No.43575216[source]▶
47% now soo a coin toss
replies(2): >>43575617 #>>43575842 #
2. layer8 ◴[03 Apr 25 21:23 UTC] No.43575617[source]▶
32% again now.
3. elicksaur ◴[03 Apr 25 21:45 UTC] No.43575842[source]▶
Note the market resolves by:

> Resolution will be via a poll of Manifold moderators. If they're split on the issue, with anywhere from 30% to 70% YES votes, it'll resolve to the proportion of YES votes.

So you should really read it as “Will >30% of Manifold moderators in 2027 think the ‘predictions seem to have been roughly correct up until that point’?”

replies(1): >>43602802 #
4. roryokane ◴[06 Apr 25 16:42 UTC] No.43602802[source]▶
That’s a misreading of the phrase “proportion of YES votes”. If 30% of judges vote YES, then only 30% – not 100% – of the prediction’s market cap is awarded to those who bet YES. The remaining 70% of the market cap is awarded to those who bet NO.

The market correctly rewards those who bet NO in such a case. Therefore, bettors have no reason to bet YES if they really think NO.