| ASHLYNN GERTH, PhD (Mille Lacs Band of Ojibwe), Grants and Policy Manager, Native BioData Consortium

The unrestrained development of artificial intelligence and machine learning (AI/ML) has increased significantly in recent years, especially in biomedical fields. Concerns have been raised about the quality of medical data being used to train AI/ML models and perpetuating racial and sociocultural biases. These concerns are not hypothetical, as numerous reports have emerged about biased AI-powered decision making in health care settings alone. Similarly, there have been concerns raised over the exploitation of genetic data for the purposes of AI/ML-driven research and development (R&D).
 
AI/ML models are valued by the quality and uniqueness of the data they are trained on. As such, lack of diversity within genomic databases has driven numerous federally-backed mass data collection initiatives aimed at minority groups, particularly Native populations. Ancient DNA unique to indigenous peoples is rare and incredibly valuable to medical researchers for potential therapeutic targets.  While efforts are marketed to improve access to personalized medicine and reduce health disparities, the underlying policies do little to enforce that outcome. Furthermore, opaque public-private partnerships held by federal agencies and universities has led many to wonder if patentable discoveries are the greater priority. However, Native peoples are not without a solution. Tribal nations claiming sovereignty over their data have the ability to circumvent historical exploitation by having the power to determine who can and cannot use their data. This position allows for a more equitable seat at the table and R&D that genuinely benefits the Native community.