Inclusive Artificial Intelligence

TL;DR

This paper proposes an alternative evaluation method for generative AIs that emphasizes inclusivity, ensuring models retain diverse knowledge for personalized responses and utility maximization, addressing limitations of existing homogeneous preference assessments.

Contribution

It introduces a new evaluation framework that promotes inclusive AI models capable of representing diverse interests and making utility-maximizing decisions.

Findings

Traditional methods favor homogeneous preferences

Proposed method enhances inclusivity and diversity in AI responses

Models retain knowledge for personalized and utility-driven decisions

Abstract

Prevailing methods for assessing and comparing generative AIs incentivize responses that serve a hypothetical representative individual. Evaluating models in these terms presumes homogeneous preferences across the population and engenders selection of agglomerative AIs, which fail to represent the diverse range of interests across individuals. We propose an alternative evaluation method that instead prioritizes inclusive AIs, which provably retain the requisite knowledge not only for subsequent response customization to particular segments of the population but also for utility-maximizing decisions.