Terminators: Terms of Service Parsing and Auditing Agents

TL;DR

Terminators addresses the challenge of making Terms of Service documents readable and auditable by decomposing the task into term extraction, verification, and accountability planning. The method uses an agentic LLM workflow to translate dense ToS into a set of verifiable, source-backed terms and context-specific accountability checks. Evaluation on the OpenAI ToS with GPT-4o demonstrates that section-wise parsing and explicit source citation reduce hallucination and increase auditability, enabling automated policy audits for regulatory or civic oversight. The framework generalizes beyond ToS, offering a query-able, computable representation of legal terms and supporting user empowerment and compliance monitoring through structured, auditable outputs.

Abstract

Terms of Service (ToS) documents are often lengthy and written in complex legal language, making them difficult for users to read and understand. To address this challenge, we propose Terminators, a modular agentic framework that leverages large language models (LLMs) to parse and audit ToS documents. Rather than treating ToS understanding as a black-box summarization problem, Terminators breaks the task down to three interpretable steps: term extraction, verification, and accountability planning. We demonstrate the effectiveness of our method on the OpenAI ToS using GPT-4o, highlighting strategies to minimize hallucinations and maximize auditability. Our results suggest that structured, agent-based LLM workflows can enhance both the usability and enforceability of complex legal documents. By translating opaque terms into actionable, verifiable components, Terminators promotes ethical use of web content by enabling greater transparency, empowering users to understand their digital rights, and supporting automated policy audits for regulatory or civic oversight.

Figures (1)

• Figure 1: Overview of Terminators.