74 INTERVIEW Turning dark data into an AI advantage Iron Mountain’s Matt Jennings explains to Lindsay James why the real bottleneck to enterprise AI isn’t the model, it’s the data feeding it. He also discusses turning decades of physical and digital records into an AI-ready foundation Industry estimates suggest as much as 80 per cent of enterprise data remains ‘dark’ – uncatalogued and unused information that is scattered across legacy systems or buried in scanned documents and email archives. Meanwhile, just under one per cent of it has made it into AI models. For Iron Mountain, that gap between what organisations hold and what they can actually use represents both a challenge and an opportunity. Matt Jennings, senior vice president for global sales and strategic partnerships within Iron Mountain’s Digital Business Unit, explains more. Iron Mountain has built its reputation on physical records storage. Why is data now such a central part of the story? Organisations still need somewhere secure and compliant to keep their most valuable information. What’s changed is what ‘valuable’ means. Every day, businesses are generating more content, but it’s scattered across systems, or boxed up and effectively invisible. In a world where speed and the use of AI now define competitive advantage, progress doesn’t come from accumulating more information – it comes from unlocking the value in what already exists. We help organisations make that information not only secure, but visible, accessible and ready for AI. Why is Microsoft Azure such an important part of that shift? The two things organisations need most right now – trustworthy AI and defensible governance – depend on having a secure, compliant cloud foundation. Iron Mountain InSight DXP runs on Microsoft Azure and uses Azure AI services like Document Intelligence and Azure OpenAI to process huge volumes of unstructured content, while inheriting Azure’s enterprise-grade security, resilience and global infrastructure. Tell us more about InSight DXP. What does it actually do? At its core, InSight DXP unifies physical and digital records into a trusted data foundation, then applies AI to make that foundation useful. We think about it in three stages: protect, connect and activate. Protect is about secure storage, chain of custody and compliance – applying retention, privacy and audit controls, whether a record is a paper file from 1995 or a document created this morning. Connect is about transforming unstructured content – digitising physical material, classifying documents and extracting the data buried inside them. And activate is where the value really shows up: AI-powered search, automated workflows and agentic AI that can handle high-volume, repetitive tasks while keeping a human in the loop for anything that needs review. “ Progress doesn’t come from accumulating more information”
RkJQdWJsaXNoZXIy NzQ1NTk=