Retrieval-Augmented Detection of Potentially Abusive Clauses in Chilean Terms of Service
2026-05-25 • Machine Learning
Machine LearningArtificial IntelligenceComputation and Language
AI summaryⓘ
The authors created a computer system to help find potentially unfair or abusive parts in online Terms of Service contracts used in Chile. Their system uses smart searching and language models that run locally on a computer, so it doesn't need big online services. They also made a collection of 100 contracts with over 10,000 clauses labeled by legal experts into different problem categories. Their tests show that combining retrieval methods with language models works better than older methods, making it easier and cheaper to check contracts for unfair rules. Finally, they offer a detailed legal labeling system and a useful design for using AI in reviewing consumer contracts.
Terms of Servicecontracts of adhesionabusive clausesconsumer lawretrieval-augmented generationlanguage modelslegal annotationcontractual imbalancegood faithAI-assisted contract review
Authors
Christoffer Loeffler, Tomás Rey Pizarro, Daniel Ignacio Miranda Vásquez, Andrea Martínez Freile
Abstract
Online Terms of Service often function as contracts of adhesion, creating asymmetries that may expose consumers to potentially abusive clauses. In Chile, assessing such clauses is legally challenging because some provisions clearly violate mandatory consumer law, whereas others depend on broader standards such as good faith and contractual imbalance. We present a retrieval-augmented generation framework for the automated detection and classification of potentially abusive clauses in Chilean Terms of Service. Designed for local execution, it combines efficient clause detection, hybrid dense--sparse retrieval, reranking, and prompt augmentation to support medium-sized open-weight language models. We also introduce the Chilean Abusive Terms of Service Extended corpus, comprising 100 contracts and 10,029 annotated clauses in 24 legally grounded categories spanning illegal, dark, and gray clauses. Experiments comparing commercial and open-weight language models, fine-tuned encoders, and traditional baselines show that retrieval-augmented prompting substantially improves performance and enables local models to approach larger cloud-based systems at lower computational and token cost. The study also contributes a refined legal annotation scheme and a practical design for AI-assisted consumer contract review.