AI projects often fail because organisations confuse having data with having AI-ready data. Data readiness depends on quality, access, ownership, governance, context and workflow fit. This article explains why leaders must assess data readiness before implementation, and why data problems are often business, process and accountability problems.