There is a tremendous opportunity for data-driven discovery across the NIH mission, including from artificial intelligence and machine learning (AI/ML) technologies. This discovery requires findable, accessible, interoperable, and reusable (FAIR) and AI/ML-ready data. Making data FAIR and AI/ML-ready requires interdisciplinary skills not typically held by biomedical and behavioral researchers. Particularly for biomedical data, AI/ML-readiness should be guided by a concern for human and clinical impact and therefore requires attention to ethical, legal, and social implications of AI/ML, such as biases in datasets, algorithms, and applications; concerns related to privacy and confidentiality; impacts on disadvantaged or marginalized groups and health disparities; and unintended, adverse social consequences of research and development.
To address these challenges, the NIH Office of Data Science Strategy was leading three NIH-wide initiatives.
Addressing the Workforce Gap in Data Governance for AI in Biomedicine
New investigators trained at the interface of information, AI, and biomedical sciences, ready to advance the field of data science for AI in biomedicine.
Ethics, Bias, and Transparency for People and Machines
Social and technical solutions for embedding ethics across the lifecycle of AI applications.
Improving the AI-readiness of Existing, IC-supported Data
Enhancing NIH data to be FAIR and AI-ready.
Meetings and Reports
- March 2024 AI PI Meeting – Details & Reports – kickoff and closeout meetings of the FY22 AI Readiness (NOT-OD-22-067) and AI Ethics (NOT-OD-22-065) and FY23 AI Readiness (NOT-OD-23-082) supplement awards
- January 2024 AI Workshop – Toward an Ethical Framework for AI in Biomedical and Behavioral Research: Transparency for Data and Model Reuse Workshop
- October 2022 AI PI Meeting – Details & Reports – kickoff and closeout meetings of the FY21 AI-readiness (NOT-OD-21-094) and AI workforce (NOT-OD-21-079) supplement awards