Ethical Dilemmas in Strategic Games
This work addresses ethical decision-making in AI and game theory, offering a formal framework for modeling dilemmas, but it is incremental as it builds on prior modalities without broad empirical validation.
The paper tackles the problem of formalizing ethical dilemmas in strategic games by introducing a new modality to capture such dilemmas, showing it cannot be defined through existing blameworthiness modalities, and providing a sound and complete axiomatization for games with perfect information.
An agent, or a coalition of agents, faces an ethical dilemma between several statements if she is forced to make a conscious choice between which of these statements will be true. This paper proposes to capture ethical dilemmas as a modality in strategic game settings with and without limit on sacrifice and for perfect and imperfect information games. The authors show that the dilemma modality cannot be defined through the earlier proposed blameworthiness modality. The main technical result is a sound and complete axiomatization of the properties of this modality with sacrifice in games with perfect information.