Data trust is the hidden reason most AI initiatives fail

A new study reveals that 90% of enterprises are deploying Enterprise GenAI at scale. However, only 34% of CISOs feel confident in their AI data security controls, resulting in a low success rate of AI initiatives. This discrepancy highlights a significant gap between AI adoption and data security readiness. Poor data governance, previously manageable, is now exposed by AI's ability to access all connected data sources. Existing security frameworks, designed for human actors, are not equipped to handle the speed and breadth of AI agents. The research shows that 70% of security leaders struggle to enforce policies on GenAI tools, and 98% face significant AI security challenges. The study encompassed surveys and interviews with 124 senior security leaders, revealing critical insights into data trust breakdown. The core argument is that data trust is essential for AI success; its absence hinders innovation and introduces risk. MIND aims to address this issue by improving data visibility, governance, and enforcement for non-human actors. The report offers a clear path for CISOs to enable AI adoption through a robust security foundation.