Research & Publishing Principles
Summary
Reproducible, verifiable, and compliant GEO research and casework.
Principles
-
Reproducibility: provide methods, data/corpus sources, prompts, and evaluation scripts or equivalents.
-
Verifiability: attach public citations or replicable experiments; include error bars or confidence notes when relevant.
-
Ethical alignment: follow OECD AI Principles and NIST AI RMF / NIST AI 600-1 risk management.
-
Legal compliance: honor copyright, privacy, and lawyer advertising rules.
-
Data governance: label source, license, collection date, and geographic limits; avoid sensitive data.
-
Evaluation: preregister metrics; report negative results; state limitations and threat models.
-
Disclosure: list funding, collaborators, and conflicts of interest.
-
Versioning & corrections: use version numbers, changelogs, and errata; timestamp updates.
-
Open publication: prefer open access; state license; provide machine-readable metadata and Schema.org.
References: OECD AI Principles; NIST AI RMF and NIST AI 600-1
Related studies: https://doi.org/10.5281/zenodo.17294708 , https://doi.org/10.5281/zenodo.17296236
Version: 1.0 • Updated: 2025-10-30