Return to Article Details Vernacularizing Taxonomies of Harm is Essential for Operationalizing Holistic AI Safety Download Download PDF