f(x)={323(4xx2)0if0x4otherwise. [25] In a particularly telling quote, Stephen Hawking, Stuart Russell, Max Tegmark, and Frank Wilczek foreshadow this stark risk: One can imagine such technology outsmarting financial markets, out-inventing human researchers, out-manipulating human leaders, and developing weapons we cannot even understand. Another example is the hunting practices of orcas (known as carousel feeding). In international relations, countries are the participants in the stag hunt. An example of norm enforcement provided by Axelrod (1986: 1100) is of a man hit in the face with a bottle for failing to support a lynching in the Jim Crow South. This same dynamic could hold true in the development of an AI Coordination Regime, where actors can decide whether to abide by the Coordination Regime or find a way to cheat. But the moral is not quite so bleak. David Hume provides a series of examples that are stag hunts. A major terrorist attack launched from Afghanistan would represent a kind of equal opportunity disaster and should make a commitment to establishing and preserving a capable state of ultimate value to all involved. Meanwhile, the harm that each actor can expect to receive from an AI Coordination Regime consists of the actors perceived likelihood that such a regime would create a harmful AI expressed as P_(h|A) (AB)for Actor A and P_(h|B) (AB)for Actor B times each actors perceived harm expressed as hA and hB. For instance if a=10, b=5, c=0, and d=2. Although the development of AI at present has not yet led to a clear and convincing military arms race (although this has been suggested to be the case[43]), the elements of the arms race literature described above suggest that AIs broad and wide-encompassing capacity can lead actors to see AI development as a threatening technological shock worth responding to with reinforcements or augmentations in ones own security perhaps through bolstering ones own AI development program. Your application of the Prisoners Dilemma (PD) game to international trade agreements raises a few very interesting and important questions for the application of game theory to real-life strategic situations. In this model, each actors incentives are not fully aligned to support mutual cooperation and thus should present worry for individuals hoping to reduce the possibility of developing a harmful AI. The United States is in the hunt, too. (required), 2023 Cornell University Powered by Edublogs Campus and running on blogs.cornell.edu, The Stag Hunt Theory and the Formation Social of Contracts, http://www.socsci.uci.edu/~bskyrms/bio/papers/StagHunt.pdf. Both nations can benefit by working together and signing the agreement. 1 The metaphors that populate game theory modelsimages such as prisoners . Based on the values that each actor assigns to their payoff variables, we can expect different coordination models (Prisoners Dilemma, Chicken, Deadlock, or Stag Hunt) to arise. [22] Julia Angwin, Jeff Larson, Surya Mattu, and Lauren Kirchner, Machine Bias, ProPublica, May 23, 2016 https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing. 2 Examples of states include the United States, Germany, China, India, Bolivia, South Africa, Brazil, Saudi Arabia, and Vietnam. See Katja Grace, John Salvatier, Allan Dafoe, Baobao Zhang, & Owain Evans, When Will AI Exceed Human Performance? In a security dilemma, each state cannot trust the other to cooperate. 75 0 obj <>stream <>stream Therefore, if it is likely that both actors perceive to be in a state of Prisoners Dilemma when deciding whether to agree on AI, strategic resources should be especially allocated to addressing this vulnerability. Game Theory 101: The Complete William Spaniel shows how to solve the Stag Hunt using pure strategy Nash equilibrium. From that moment on, the tenuous bonds keeping together the larger band of weary, untrusting hunters will break and the stag will be lost. LTgC9Nif The area of international relations theory that is most characterized by overt metaphorical imagery is that of game theory.Although the imagery of game theory would suggest that the games were outgrowths of metaphorical thinking, the origins of game theory actually are to be found in the area of mathematics. As described in the previous section, this arms race dynamic is particularly worrisome due to the existential risks that arise from AIs development and call for appropriate measures to mitigate it. Huntington[37] makes a distinction between qualitative arms races (where technological developments radically transform the nature of a countrys military capabilities) and quantitative arms races (where competition is driven by the sheer size of an actors arsenal). The second technology revolution caused World War II. Table 8. Meanwhile, the harm that each actor can expect to receive from an AI Coordination Regime consists of both the likelihood that the actor themselves will develop a harmful AI times that harm, as well as the expected harm of their opponent developing a harmful AI. Charisma unifies people supposedly because people aim to be as successful as the leader. The reason is because the traditional PD game does not fully capture the strategic options and considerations available to each player. An approximation of a Stag Hunt in international relations would be an international treaty such as the Paris Climate Accords, where the protective benefits of environmental regulation from the harms of climate change (in theory) outweigh the benefits of economic gain from defecting. In chapter 6 of Man, the State, and War, precursor of the anarchical view of international relations, and an extension of the stag-hunt example: This can be facilitated, for example, by a state leader publicly and dramatically expressing understanding of danger and willingness to negotiate with other states to achieve this. 0000016685 00000 n [5] Stuart Armstrong, Nick Bostrom, & Carl Shulman, Racing to the precipice: a model of artificial intelligence development, AI and Society 31, 2(2016): 201206. For example, most land disputes, like the ongoing Chinese andJapanesedisputeover the Senkaku Islands, must be resolved bycompromisingin other areas of policy in order to achieve the goal. Orcas cooperatively corral large schools of fish to the surface and stun them by hitting them with their tails. As such, Chicken scenarios are unlikely to greatly affect AI coordination strategies but are still important to consider as a possibility nonetheless. [41] AI, being a dual-use technology, does not lend itself to unambiguously defensive (or otherwise benign) investments. Here, we assume that the harm of an AI-related catastrophe would be evenly distributed amongst actors. Here, I also examine the main agenda of this paper: to better understand and begin outlining strategies to maximize coordination in AI development, despite relevant actors varying and uncertain preferences for coordination. As stated, which model (Prisoners Dilemma, Chicken, Deadlock, or Stag Hunt) you think accurately depicts the AI Coordination Problem (and which resulting policies should be pursued) depends on the structure of payoffs to cooperating or defecting. [7] Aumann concluded that in this game "agreement has no effect, one way or the other." I will apply them to IR and give an example for each. They will be tempted to use the prospect of negotiations with the Taliban and the upcoming election season to score quick points at their rivals expense, foregoing the kinds of political cooperation that have held the country together until now. Weiss, Uri, and Joseph Agassi. What are the two exceptions to the ban on the use of force in the UN Charter? Explain how the 'Responsibility to Protect' norm tries to provide a compromise between the UN Charter's principle of non-interference (state sovereignty) and the UN genocide convention. If both choose to row they can successfully move the boat. In this section, I survey the relevant background of AI development and coordination by summarizing the literature on the expected benefits and harms from developing AI and what actors are relevant in an international safety context. Overall, the errors overstated the companys net income by 40%. As such, it will be useful to consider each model using a traditional normal-form game setup as seen in Table 1. [11] This Article conceptualizes a stag hunt in which the participants are countries that host extractive companies on their stock exchanges, including the U.S., Canada, the United Kingdom, the Member States . 695 0 obj Despite the damage it could cause, the impulse to go it alone has never been far off, given the profound uncertainties that define the politics of any war-torn country. In biology many circumstances that have been described as prisoner's dilemma might also be interpreted as a stag hunt, depending on how fitness is calculated. The real peril of a hasty withdrawal of U.S. troops from Afghanistan, though, can best be understood in political, not military, terms. In this article, we employ a class of symmetric, ordinal 2 2 games including the frequently studied Prisoner's Dilemma, Chicken, and Stag Hunt to model the stability of the social contract in the face of catastrophic changes in social relations. 8,H7kcn1qepa0y|@. Scholars of civil war have argued, for example, that peacekeepers can preserve lasting cease-fires by enabling warring parties to cooperate with the knowledge that their security will be guaranteed by a third party. Leanna Litsch, Kabul Security Force Public Affairs. 0000004367 00000 n 'War appears to be as old as mankind, but peace is a modern invention'. trailer In addition to leadership, the formation of a small but successful group is also likely to influence group dynamics. Payoff matrix for simulated Prisoners Dilemma. Like the hunters in the woods, Afghanistans political elites have a great deal, at least theoretically, to gain from sticking together. [15] Sam Byford, AlphaGo beats Lee Se-dol again to take Google DeepMind Challenge series, The Verge, March 12, 2016, https://www.theverge.com/2016/3/12/11210650/alphago-deepmind-go-match-3-result. This means that it remains in U.S. interests to stay in the hunt for now, because, if the game theorists are right, that may actually be the best path to bringing our troops home for good. One final strategy that a safety-maximizing actor can employ in order to maximize chances for cooperation is to change the type of game that exists by using strategies or policies to affect the payoff variables in play. The question becomes, why dont they always cheat? a This table contains an ordinal representation of a payoff matrix for a Chicken game. Together, these elements in the arms control literature suggest that there may be potential for states as untrusting, rational actors existing in a state of international anarchy to coordinate on AI development in order to reduce future potential global harms.