About OpenPolicy AI
Mission
OpenPolicy AI provides transparent, data-driven evaluation of US politicians' alignment with internationally recognized AI principles established by the OECD. Our goal is to make AI policy understanding clear, accessible, and transparent.
Methodology
We analyze politicians' voting records, bill sponsorships, and public statements using a multi-signal evidence framework to evaluate alignment with 5 core OECD AI principles:
- Inclusive Growth, Sustainable Development & Well-being
- Human-Centered Values & Fairness
- Transparency & Explainability
- Robustness, Security & Safety
- Accountability
How It Works
1. Evidence Collection — We collect floor votes, bill sponsorships, committee statements, and press releases from public sources.
2. AI Classification — Claude AI classifies each evidence item for AI relevance, bill direction, and extracts structured claims.
3. Deterministic Scoring — All scores are calculated using transparent, deterministic math. The AI never produces scores directly.
Transparency
All scores are accompanied by confidence levels. Every score is traceable to specific evidence items, including source text and extracted claims. We display our full methodology to ensure accountability.
Disclaimer
This platform is for informational purposes only. Scores are based on computational analysis of public data and should not be taken as definitive assessments. Always verify with primary sources.
