VALORIALAI Ethics and Compliance

Resources

Methodology and publications.

Valorial publishes its methodology openly. If the FRIA framework holds up, it should withstand public scrutiny.

Methodology Paper

From Safeguards to Fundamental Rights: Applying IFC Proportionality Methodology to Fundamental Rights Impact Assessments under EU AI Act Article 27

SSRN Preprint · 2026

Available

Article 27 mandates FRIA content but not methodology. This paper argues that existing questionnaire-based approaches produce documentation without analytical rigour. IFC PS1 proportionality methodology — developed over 30 years in development finance — transfers directly to FRIA contexts, producing scored, weighted outputs superior to existing guidance (DIHR/ECNL, ALIGNER, Bertaina et al.). The paper presents a full FRIA workbook methodology and three worked examples.

Read Paper →

Blog Post

Grievance Mechanisms for AI: What the EU AI Act Requires and Why No One Is Building It Right

Valorial · 2026

Coming soon

Every compliance platform focuses on documentation. None of them design the operational process for what happens when your AI system harms someone. Article 26(1)(d) requires accessible means for affected persons to raise concerns — and this piece argues that IFC-style grievance mechanism design is the only methodology robust enough to meet that standard.

OECD Blog

Grievance and Redress Mechanisms for AI: Applying Development Finance Practice to Automated Decision Systems

OECD · 2026

Coming soon

A cross-sector analysis of how grievance mechanism design principles from OECD Responsible Business Conduct frameworks and IFC Performance Standards can be applied to AI-affected persons in regulated deployment contexts.

Stay current

EU AI Act deadlines move. We track them.

Regulatory updates, methodology notes, and compliance briefings.

Subscribe →