I do think there's still more thinking to be done here, but, since I recorded the episode, Alexis Carlier and Tom Davidson have actually done some good work in response to Hanson's critique. I was pretty persuaded of their conclusion:
There are similarities between the AI alignment and principal-agent problems, suggesting that PAL could teach us about AI risk. However, the situations economists have studied are very different to those discussed by proponents of AI risk, meaning that findings from PAL don’t transfer easily to this context. There are a few
Do you still think that Robin Hanson's critique of Christiano's scenario is worth exploring in more detail?
I do think there's still more thinking to be done here, but, since I recorded the episode, Alexis Carlier and Tom Davidson have actually done some good work in response to Hanson's critique. I was pretty persuaded of their conclusion:
... (read more)