
0:00
28:44
There is a lot of talk about AI alignment, which is getting machines to act in our best interests. But how do we go about doing this? The philosophical secret to this conundrum is values. But how do we determine what human values are? This problem needs to be solved before we program computers to do this or that on our behalf. In this two-part series, I discuss what values are and look at evidence on human values coming from evolutionary theory, literature reviews and survey research. The talk rounds out with the contentious debate on whether human values are relative or universal.
More episodes from "Save the World with Jay Friedenberg"



Don't miss an episode of “Save the World with Jay Friedenberg” and subscribe to it in the GetPodcast app.








