Responsible LLM Deployment for High-Stake Decisions by Decentralized Technologies and Human-AI Interactions

TL;DR

This paper presents a framework for responsible deployment of LLMs in high-stakes decision-making, emphasizing human oversight, decentralized records, and rigorous evaluation to ensure security, accountability, and reliability.

Contribution

It introduces a comprehensive framework combining human-AI collaboration, decentralized technology integration, and iterative evaluation for safe LLM deployment in critical domains.

Findings

Effective human-AI collaboration improves decision quality.

Decentralized records enhance security and accountability.

Framework supports responsible financial decision-making.

Abstract

High-stakes decision domains are increasingly exploring the potential of Large Language Models (LLMs) for complex decision-making tasks. However, LLM deployment in real-world settings presents challenges in data security, evaluation of its capabilities outside controlled environments, and accountability attribution in the event of adversarial decisions. This paper proposes a framework for responsible deployment of LLM-based decision-support systems through active human involvement. It integrates interactive collaboration between human experts and developers through multiple iterations at the pre-deployment stage to assess the uncertain samples and judge the stability of the explanation provided by post-hoc XAI techniques. Local LLM deployment within organizations and decentralized technologies, such as Blockchain and IPFS, are proposed to create immutable records of LLM activities for automated auditing to enhance security and trace back accountability. It was tested on Bert-large-uncased, Mistral, and LLaMA 2 and 3 models to assess the capability to support responsible financial decisions on business lending.