Agreed. We think our human-AI setting is a useful model of alignment in the limit case, but not really so in the transient case. (For the reason you point out.)
Agreed. We think our human-AI setting is a useful model of alignment in the limit case, but not really so in the transient case. (For the reason you point out.)