---
pretty_name: ADGL Reference Cases and Governance Profiles
language:
- en
license: apache-2.0
size_categories:
- n<1K
tags:
- ai-governance
- data-governance
- responsible-ai
- policy
- conformance
- audit
- provenance
- enterprise-ai
configs:
- config_name: reference_cases
  data_files:
  - split: train
    path: data/reference_cases.jsonl
  default: true
- config_name: governance_profiles
  data_files:
  - split: train
    path: data/profiles.jsonl
---

# ADGL Reference Cases and Governance Profiles

This Dataset repository accompanies the **AI Data Governance Layer (ADGL)** public research project.

ADGL models **Knowledge Governance → Analysis Governance → Consequence Governance**, with `INFORM`, `DECIDE`, and `ACT` as principal consequence dispositions and Audit + Provenance spanning the complete governance trajectory.

## Configurations

- `reference_cases`: eight structured reference cases with policy, input fixture, and expected result.
- `governance_profiles`: five illustrative, non-normative domain profiles.

## Current integrity status

Toolkit version: **0.5.3**

The corresponding reference toolkit reports **25/25 published normative checks** plus **11/11 candidate execution-integrity checks**.

This dataset is for standards research, independent implementation, conformance experimentation, governance benchmark development, and demonstrations. It is not a training corpus, legal-compliance certification dataset, or production authorization database.

## Licensing

This Hugging Face Dataset package contains executable fixtures, schemas, and illustrative profiles released under **Apache License 2.0**. The separately published ADGL paper/specification uses its own documentation license.

Creator: **GBSN Research**. For project contact, go to **www.gbsnresearch.com** and use **Contacts**.