E-discovery platforms have gotten great at narrowing millions of documents down to manageable sets. But what happens next \u2014 the grueling work of extracting facts, organizing them, and building a reliable case narrative \u2014 has remained largely manual. In this episode of LawNext, host Bob Ambrogi talks with Daniel Lord-Doyle, cofounder and CEO of Mary [\u2026]<\/p>\n
E-discovery platforms have gotten great at narrowing millions of documents down to manageable sets. But what happens next \u2014 the grueling work of extracting facts, organizing them, and building a reliable case narrative \u2014 has remained largely manual. In this episode of LawNext, host Bob Ambrogi talks with Daniel Lord-Doyle<\/a>, cofounder and CEO of Mary Technology<\/a>, about the Australian startup\u2019s bet that \u201cfact management\u201d is the missing layer in litigation technology.<\/p>\n Mary\u2019s approach is distinct from the large AI platforms that store documents as embeddings in vector databases. Instead, the company extracts every individual fact, enriches it with metadata, and links it directly back to its source \u2014 creating what Lord-Doyle calls a verifiable chain from work product to evidence. He makes a compelling case that in litigation, where fault tolerance is low and the stakes are high, the nuance lost by compression-based AI systems is exactly what matters most.<\/p>\n