Am I wrong to feel that your article considers alignment as a safety test run by an AI on the next generation of AIs ?
I feel like, when most people talk about automating Alignment research, they hope to have a country of geniuses in a datacenter that would build the entire theory from the ground up, invent the necessary new concepts and build a formalism around intellodynamics that would allow any kind of research to then be pursued in a safe manner.
As estimated by some, this would take around centuries of work for humans that could be compressed into ye...
Don't you think that the "mysterious" progress on morality comes from the fact that the incentives for it and the easyness of it both increased via technology ?
1) the incentives
I mean, beeing nice has always had the side effect of earning trust. I don't mean that selfless actions do not exist. For instance, long term vegans generally do not gain anything in their choice to refuse animal consumption. But one must admit that beeing nice in public has some benefit for a person.
With the appearance of video technology, showing how nice one is has become more r... (read more)