The Future of Data Workshop, held at the University of Missouri on October 1st and 2nd, examined the growing risk that important federal data and reporting systems may be reduced, substantially altered, or discontinued. Many academic, policy, and public sector projects depend on federal information infrastructures that have been built over decades. When those systems change or disappear, the problem is not simply the loss of a dataset. Researchers also risk losing continuity over time, institutional knowledge, established data generating procedures, and the ability to make reliable comparisons across years. The workshop therefore focused on how these valuable public data infrastructures might be preserved, reconstructed, or supplemented when their traditional federal sources become uncertain.
Artificial intelligence and machine learning were central to the discussion. Participants considered whether AI could help reconstruct parts of the processes through which federal data are generated, including information collection, source evaluation, classification, synthesis, and reporting. A major theme was that effective use of AI requires much more than feeding historical documents into a model. Researchers first need to understand the human expertise, institutional procedures, and professional judgments that produced those data in the first place.
To make that conversation possible, the workshop brought together academic researchers with former federal practitioners who had direct experience producing and overseeing important government reporting systems. Participants included senior former State Department officials and longtime Human Rights Report practitioners, including Roberta Cohen, former Deputy Assistant Secretary of State for Human Rights, and Stephen Eisenbraun, former Coordinator of Human Rights Reports and Editor in Chief of the State Department's Country Reports on Human Rights Practices. Other former State Department officials and HRR editors also participated, providing researchers with a rare opportunity to understand the institutional and human processes behind federal data production rather than looking only at the final published reports.
The State Department's Country Reports on Human Rights Practices served as an important case through which the group explored these broader questions. For decades, academic researchers have relied on these reports to understand human rights conditions and to construct quantitative measures such as the Political Terror Scale. Participants discussed what happens when a long-standing federal reporting system changes, which elements of the original data generating process need to be preserved, and whether AI and machine learning can help reconstruct parts of that process.
Former State Department practitioners also participated in a special session with undergraduate students. They shared their experiences working in the State Department and the U.S. Foreign Service, discussed different career paths in diplomacy and human rights, and offered students practical advice based on their own professional experiences. This gave students a valuable opportunity to engage directly with people who had spent many years working on human rights and U.S. foreign policy.
More broadly, the workshop demonstrated that preserving federal data is not simply an archival challenge. It also requires preserving expertise, procedures, judgment, and institutional memory. The discussions highlighted how collaboration among researchers, former federal practitioners, and AI experts may offer new ways to maintain the continuity and public value of federal data when traditional information infrastructures are at risk.
.