AXRP
Paper authors talk AI safety: METR time horizons, effective compute — numbers come with sources, two to three hours an episode
2:40:40AI 2027's author: aligning the next generation of AI takes AI a thousand times less effort than humans
In AI 2027, humans aligning superintelligence face overwhelming odds, yet agent 4 aligning the next generation succeeds almost on the first try — Eli admits the asymmetry is real, and that AI doing it itself saves at least a thousand times the cognition.
2:32:07Program Equilibrium: Trading Code Cooperates More Than Trading People, but Coordination Is Extremely Hard
Program equilibrium lets two parties read each other's source code and achieve cooperation via Löb's theorem; but cooperation depends on nearly identical code or shared randomness, and in reality coordination costs are extremely high.
2:05:34Giving AI Property Rights Is How You Stop It From Flipping the Table
If AI owns property, it has less incentive to destroy the property system itself — including taking everything humans own. This isn't moral concern; it's risk hedging.