A form of AI Governance that involves identify potential harms and applying specific fixes to address those harms