The principle "Global AI Governance Initiative" has mentioned the topic "safety" in the following places: