Open research resources for Saudi Arabic legal NLP, information retrieval, Retrieval-Augmented Generation, and reliable language model evaluation.
Saudi Arabic Laws is a research initiative focused on developing structured, reusable resources for computational research on Saudi legal text.
Our work emphasizes evidence-grounded language models, legal information retrieval, dataset quality, and reproducible evaluation using material derived from publicly accessible official Saudi legal sources.
A structured, article-level Arabic legal corpus designed for retrieval, RAG, grounded generation, and legal LLM evaluation.
| Legal articles | 22,593 |
| Source documents | 578 |
| Sources | BOE + MOJ |
| Language | Arabic |
| Granularity | Legal article |
| Release | v1.0.0 |
| License | CC BY-NC-ND 4.0 |
Explore the dataset:
Saudi Arabic Laws and Regulations Corpus
Archival release:
Zenodo — Version 1.0.0
Our resources are designed to support research in areas including:
The corpus integrates legal material collected from publicly accessible official Saudi government sources:
The released datasets preserve source-specific structure and metadata where appropriate.
The team develops and maintains the research resources published through this organization collectively.
For academic use of the Saudi Arabic Laws and Regulations Corpus, please cite the archived Zenodo release:
Almunyah, A., Almehmadi, A., Almaashi, F., Sirdar, H., & Alamri, S. (2026).
Saudi Arabic Laws and Regulations Corpus (Version 1.0.0) [Dataset]. Zenodo.
https://doi.org/10.5281/zenodo.21265180
Resources published through this organization are intended for research and educational use in legal NLP and reliable AI.
They should not be interpreted as official legal publications, authoritative legal interpretations, or substitutes for professional legal advice. For legally consequential use, information should be verified against the relevant current official Saudi government source.