Evaluating the Risks of Preventive Attack in the Race for Advanced AI

Will the race for advanced artificial intelligence make war more likely? If technology futurists are right that the advent of artificial general intelligence (AGI) will radically alter the global balance of power, could AGI development tempt states to resort to war to secure those advantages for themselves or to deny them to their rivals? The authors developed a framework of variables that shape pressures for preventive military action and considered how these variables apply to the development of AGI. The authors analyzed three scenarios for preventive military action. The findings suggest that, in absolute terms, the probability of war is low, but preventive attacks appear relatively more likely to occur in an attempt to preserve a monopoly on AGI than in an attempt to prevent one.