Efforts to curb 'shadow AI' use by employees may inadvertently increase security risks by driving usage underground and reducing visibility for security teams. The widespread adoption of unapproved AI tools necessitates a shift in strategy beyond simple blacklisting.