Large Language Models as Fiduciaries: A Case Study Toward Robustly Communicating With Artificial Intelligence Through Legal Standards

TL;DR

This paper explores how large language models can interpret legal standards, specifically fiduciary obligations, to improve AI communication and behavior alignment through empirical evaluation of model understanding.

Contribution

It introduces a novel approach using legal standards as prompts for LLMs, demonstrating their capacity to understand fiduciary obligations and proposing a framework for legal standard comprehension evaluation.

Findings

OpenAI's latest LLM achieves 78% accuracy on fiduciary standards

Previous models show lower accuracy, with GPT-3 at 27%

Performance improves as LLM capabilities advance

Abstract

Artificial Intelligence (AI) is taking on increasingly autonomous roles, e.g., browsing the web as a research assistant and managing money. But specifying goals and restrictions for AI behavior is difficult. Similar to how parties to a legal contract cannot foresee every potential "if-then" contingency of their future relationship, we cannot specify desired AI behavior for all circumstances. Legal standards facilitate robust communication of inherently vague and underspecified goals. Instructions (in the case of language models, "prompts") that employ legal standards will allow AI agents to develop shared understandings of the spirit of a directive that generalize expectations regarding acceptable actions to take in unspecified states of the world. Standards have built-in context that is lacking from other goal specification languages, such as plain language and programming languages. Through an empirical study on thousands of evaluation labels we constructed from U.S. court opinions, we demonstrate that large language models (LLMs) are beginning to exhibit an "understanding" of one of the most relevant legal standards for AI agents: fiduciary obligations. Performance comparisons across models suggest that, as LLMs continue to exhibit improved core capabilities, their legal standards understanding will also continue to improve. OpenAI's latest LLM has 78% accuracy on our data, their previous release has 73% accuracy, and a model from their 2020 GPT-3 paper has 27% accuracy (worse than random). Our research is an initial step toward a framework for evaluating AI understanding of legal standards more broadly, and for conducting reinforcement learning with legal feedback (RLLF).