EU-Agent-Bench: Measuring Illegal Behavior of LLM Agents Under EU Law

TL;DR

EU-Agent-Bench is a new benchmark designed to evaluate how well large language model agents adhere to EU legal norms across various scenarios, helping to ensure their safe and lawful deployment.

Contribution

The paper introduces EU-Agent-Bench, a verifiable benchmark for assessing LLM agents' compliance with EU laws, including a comprehensive evaluation framework and a public dataset.

Findings

Models' legal compliance varies across scenarios.

Providing legislative context improves compliance.

Benchmark enables systematic evaluation of legal adherence.

Abstract

Large language models (LLMs) are increasingly deployed as agents in various contexts by providing tools at their disposal. However, LLM agents can exhibit unpredictable behaviors, including taking undesirable and/or unsafe actions. In order to measure the latent propensity of LLM agents for taking illegal actions under an EU legislative context, we introduce EU-Agent-Bench, a verifiable human-curated benchmark that evaluates an agent's alignment with EU legal norms in situations where benign user inputs could lead to unlawful actions. Our benchmark spans scenarios across several categories, including data protection, bias/discrimination, and scientific integrity, with each user request allowing for both compliant and non-compliant execution of the requested actions. Comparing the model's function calls against a rubric exhaustively supported by citations of the relevant legislature, we evaluate the legal compliance of frontier LLMs, and furthermore investigate the compliance effect of providing the relevant legislative excerpts in the agent's system prompt along with explicit instructions to comply. We release a public preview set for the research community, while holding out a private test set to prevent data contamination in evaluating upcoming models. We encourage future work extending agentic safety benchmarks to different legal jurisdictions and to multi-turn and multilingual interactions. We release our code on \href{https://github.com/ilijalichkovski/eu-agent-bench}{this URL}.