Towards Foundation-model-based Multiagent System to Accelerate AI for Social Impact

TL;DR

This paper proposes a foundation-model-based multi-agent system to streamline AI for social impact, reducing resource demands and enhancing scalability through a human-in-the-loop approach.

Contribution

It introduces a novel meta-level multi-agent system leveraging foundation models to accelerate AI4SI development and address resource and expertise limitations.

Findings

Framework for resource allocation in AI4SI pipeline

Incorporation of foundation models for problem-solving

Emphasis on ethical and human-in-the-loop considerations

Abstract

AI for social impact (AI4SI) offers significant potential for addressing complex societal challenges in areas such as public health, agriculture, education, conservation, and public safety. However, existing AI4SI research is often labor-intensive and resource-demanding, limiting its accessibility and scalability; the standard approach is to design a (base-level) system tailored to a specific AI4SI problem. We propose the development of a novel meta-level multi-agent system designed to accelerate the development of such base-level systems, thereby reducing the computational cost and the burden on social impact domain experts and AI researchers. Leveraging advancements in foundation models and large language models, our proposed approach focuses on resource allocation problems providing help across the full AI4SI pipeline from problem formulation over solution design to impact evaluation. We highlight the ethical considerations and challenges inherent in deploying such systems and emphasize the importance of a human-in-the-loop approach to ensure the responsible and effective application of AI systems.