has anyone tested how good the major llms are at predicting each other? particularly wrt moral dilemmas and scissor questions
hmm also maybe setting them up to play games like nomic against each other? https://content.cooperate.com/post/nomic/
mimi10v3