AI-Ready Data: From Public Data to Public Value (UNGA)

Countries and institutions possess enormous volumes of data. Yet much of it remains fragmented, poorly described, difficult to access and unusable by people or AI systems. Across the world, public data holds the potential to improve healthcare, education, agriculture, livelihoods, public administration and economic growth. However, digitisation alone does not make data useful. Institutions must identify high-value datasets, establish standards, resolve governance and privacy questions, enable responsible access and build the capability to maintain and use data over time. Through real-world data-unlock journeys and facilitated working sessions, participants will explore what it takes to move a dataset from simply existing to creating measurable public value. The discussion will focus on the institutional and technical foundations often missing from conventional AI conversations: stewardship, interoperability, governance, accountable ownership, responsible access and AI readiness. The workshop will introduce Data DHARA, a seven-stage lifecycle for making priority datasets usable: Identified → Described → Interoperable → Catalogued → Governed → Accessible → AI-ready The opportunity is to help countries activate their own data assets while retaining the capability to govern how those assets are accessed and used. Agenda Outcome The workshop will invite participants to identify a high-value dataset or public problem they can help unlock, and the role they can play as a government, funder, multilateral, technology organisation or implementation partner. The ambition is not simply to release more data or build more AI applications. It is to strengthen the institutional capability required for countries to govern their data, activate it responsibly and convert it into lasting economic and social value.