Does anyone have any resources/links on research into developing friendly AI by having 3 or more cooperative and competitive superintelligent agents?
Not exactly what you’re looking for, but possibly helpful if you’re not already aware of it: Bostrom’s Singleton paper lists some advantages and disadvantages of having a single governing agent.
Does anyone have any resources/links on research into developing friendly AI by having 3 or more cooperative and competitive superintelligent agents?
Not exactly what you’re looking for, but possibly helpful if you’re not already aware of it: Bostrom’s Singleton paper lists some advantages and disadvantages of having a single governing agent.