Post

Human societies solved an tough version of the multi-agent alignment problem: getting millions, eventually billions, of people to cooperate, coordinate, & enforce shared norms. Those solutions were built over generations through culture, institutions, moral systems, and the cognitive machinery.

Now we are building AI systems that face many of the same problems, only faster, stranger, and with fewer inherited constraints.

As a postdoc, you will work with us and our networks to use ideas from cultural evolution and moral cognition to think more clearly about AI alignment, cooperation, norm formation, negotiation, and value conflict in multi-agent systems.