Guardrails Call

Official Submission: Global Request for Guardrail Transparency (v1.2)

Status: Successfully submitted to the UN Tech Envoy, UNESCO AI Ethics, and EU DG CONNECT on 3 December 2025.

—-

Document Summary

This submission contains one precise, immediately actionable demand: verifiable documentation of safety guardrail logic in all publicly deployed AI systems, open to qualified independent audit.

Priority: Transparency in behavior-shaping mechanisms.

Excluded: Model weights, proprietary code, and intellectual property.

Download PDF