Arnaud Fickinger

Papers

Total Citations

H-Index

About

Arnaud Fickinger is a researcher whose work lies at the intersection of artificial intelligence, game theory, and social choice, with a particular focus on ensuring that AI systems can serve multiple human stakeholders equitably. His primary research area centers on **multi-principal assistance games (MPAGs)** , a novel framework he introduced to address the challenge of a single AI agent acting on behalf of several humans who may hold conflicting preferences. Fickinger’s major contribution is the formal definition of MPAGs and the proposal of “collegial mechanisms” that circumvent classic impossibility results in social choice theory, such as Gibbard’s theorem, by leveraging sufficiently collegial preference inference. This work represents a significant step toward designing AI systems that are not only helpful but also fair and democratic in multi-stakeholder environments. While his most-cited paper, “Multi-Principal Assistance Games: Definition and Collegial Mechanisms” (2020), has garnered 4 citations, its conceptual novelty is notable for laying the groundwork for a new subfield in beneficial AI. Fickinger’s research is particularly relevant for students and researchers interested in the alignment problem, cooperative AI, and the ethical deployment of autonomous systems in human societies.