AI brief
A LessWrong post argues that a public agreement to punish any advantage gained from defecting more than proportionately could reinforce other agreements without needing additional mechanisms.
Why it matters: It suggests a low-cost way to reduce loss-of-control risks in AI by strengthening cooperation agreements.
Written by AI from LessWrong's published text.
Source
“Tl;dr - Public agreement that any advantage gained from defecting should be punished more than would be proportionate to any advantages gained is important.”
Interpretation or community commentary rather than straight reporting.
Read original story ↗