habryka comments on Mitigating extreme AI risks amid rapid progress [Linkpost]

[habryka]

Pretty plausible! I am not that intimately familiar with academic norms.