Cool work! Did you all test to see what happens if the identity of the other agents isn’t mentioned at all? And how about if you say „perfectly rational agents“? Wondering how much of this is just prompt sensitivity to questions that might be on a game theory exam.
Also, testing „perfectly rational agents“ would allow you to see if it’s AI preferentially cooperating with each other or superrat in general that’s driving this.
Cool work! Did you all test to see what happens if the identity of the other agents isn’t mentioned at all? And how about if you say „perfectly rational agents“? Wondering how much of this is just prompt sensitivity to questions that might be on a game theory exam.
Also, testing „perfectly rational agents“ would allow you to see if it’s AI preferentially cooperating with each other or superrat in general that’s driving this.
It would also be interesting to prompt the model with “the other players are AI agents much more capable than you”.