EthicsTrustworthinessAccountabilitycs.AI
Using experimental game theory to transit human values to ethical AI
Yijia Wang, Yan Wan, Zhijian Wang
November 16, 2017
arXiv.org
4 citations
Key Takeaways
-
Game theory can bridge philosophy and AI practice: The paper demonstrates how experimental game theory provides a concrete mathematical method to connect abstract ethical theories (like Kantarianism and Utilitarianism) with practical AI industry standards (IEEE P7000 series), making philosophical concepts actionable for AI developers.
-
Human values can be extracted from real behavior: By using experimental games like trust games, researchers can observe how humans actually behave in ethically-relevant situations rather than relying solely on theoretical assumptions, providing empirical data to train ethical AI systems.
-
AI ethics can be tested experimentally: The paper proposes using iterated Prisoner's Dilemma experiments as a method to evaluate whether specific AI algorithms behave ethically, offering a testable framework rather than relying only on theoretical analysis or subjective judgment.
-
Addresses a critical gap in AI ethics: The approach tackles the practical challenge of translating high-level ethical principles into concrete technical standards that AI developers can actually implement and measure, moving beyond abstract discussions to implementable solutions.
Abstract
Knowing the reflection of game theory and ethics, we develop a mathematical representation to bridge the gap between the concepts in moral philosophy (e.g., Kantian and Utilitarian) and AI ethics industry technology standard (e.g., IEEE P7000 standard series for Ethical AI). As an application, we demonstrate how human value can be obtained from the experimental game theory (e.g., trust game experiment) so as to build an ethical AI. Moreover, an approach to test the ethics (rightness or wrongness) of a given AI algorithm by using an iterated Prisoner's Dilemma Game experiment is discussed as an example. Compared with existing mathematical frameworks and testing method on AI ethics technology, the advantages of the proposed approach are analyzed.