Can We Trust AI Agents? A Case Study of an LLM-Based Multi-Agent System for Ethical AI

2024-10-25

AI Summary

This work addresses the trustworthiness bottleneck of large language model (LLM)-based multi-agent systems in AI ethics practice. We propose a novel framework integrating Design Science Research (DSR) with a multi-agent debate paradigm. The approach employs role-based LLM specialization, structured prompting, and iterative deliberation mechanisms to enable end-to-end analysis of real-world ethical incidents and automatic generation of compliant code and documentation. To our knowledge, this is the first work to introduce debate-driven collaboration for operationalizing AI ethics, covering critical dimensions including bias detection, GDPR compliance, and the EU AI Act. Empirical evaluation demonstrates that our system generates, on average, 2,000 lines of high-quality, ethically aligned code per case—compared to a baseline of only 80 lines—significantly improving the completeness and executability of ethical responses. However, challenges remain in engineering-level system integration and scalability.

Technology Category

• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy
• Machine Learning: Ethics, Bias, and Fairness

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Search and Retrieval-Augmented AI: Agentic search

Abstract

AI-based systems, including Large Language Models (LLM), impact millions by supporting diverse tasks but face issues like misinformation, bias, and misuse. AI ethics is crucial as new technologies and concerns emerge, but objective, practical guidance remains debated. This study examines the use of LLMs for AI ethics in practice, assessing how LLM trustworthiness-enhancing techniques affect software development in this context. Using the Design Science Research (DSR) method, we identify techniques for LLM trustworthiness: multi-agents, distinct roles, structured communication, and multiple rounds of debate. We design a multi-agent prototype LLM-MAS, where agents engage in structured discussions on real-world AI ethics issues from the AI Incident Database. We evaluate the prototype across three case scenarios using thematic analysis, hierarchical clustering, comparative (baseline) studies, and running source code. The system generates approximately 2,000 lines of code per case, compared to only 80 lines in baseline trials. Discussions reveal terms like bias detection, transparency, accountability, user consent, GDPR compliance, fairness evaluation, and EU AI Act compliance, showing this prototype ability to generate extensive source code and documentation addressing often overlooked AI ethics issues. However, practical challenges in source code integration and dependency management may limit its use by practitioners.

Problem

• Assessing LLM trustworthiness in AI ethics applications
• Evaluating multi-agent systems for ethical AI decision-making
• Addressing practical challenges in AI ethics code implementation

Innovation

• Multi-agent system with structured communication
• LLM trustworthiness via debate and roles
• Generates code for AI ethics compliance