We study a program game version of the Prisoner's Dilemma, i.e., a two-player game in which each player submits a computer program, the programs are given read access to each other's source code and then choose whether to cooperate or defect. Prior work has introduced various programs that form cooperative equilibria against themselves in this game. For example, the ϵ-grounded Fair Bot cooperates with probability ϵ and with the remaining probability runs its opponent's program and copies its action. If both players submit this program, then this is a Nash equilibrium in which both players cooperate. Others have proposed cooperative equilibria based on proof-based Fair Bots, which cooperate if they can prove that the opponent cooperates (and defect otherwise). We here show that these different programs are compatible with each other. For example, if one player submits ϵ-grounded Fair Bot and the other submits a proof-based Fair Bot, then this is also a cooperative equilibrium of the program game version of the Prisoner's Dilemma.
osgt most naturally can make ai cooperate against humans if the humans don't also understand how to formally bind themselves usefully and safely and reliably. however, humans are able to make fairly strongly binding commitments to use strategies that condition on how others choose strategies, and there are a bunch of less exact strategy inference papers I could go hunt down that are pretty interesting.