publications.

We assess empathetic LLMs from a perfectionist framework, arguing that the way these models are being used as AI companions will likely impoverish users’ social capacities required for virtue friendships.

I argue that the demands of beneficence are stronger than we think, because giving to charity tends to affect our distant-future interests, rather than our present ones. 

I argue that MEC is stuck between a rock and a hard place: it is either too demanding, or it fails to make room for supererogatory acts. I offer a solution which gets it out of this conundrum. 

I argue that we should save the many over the few, just in case Taurek is wrong and the numbers do count.