The Artificial Intelligence Review Framework documents the Yale AI Governance Committee’s principles for evaluating proposed AI systems to ensure their appropriate use. It establishes a core framework for trustworthy AI, addressing fairness, transparency, data privacy, and human oversight. The framework considers risk, applying more rigorous scrutiny to high-risk activities like clinical care and hiring. The document includes a table of principles with key considerations as well as a table of illustrative examples applied to the AI tool life cycle. Taken together, the Framework is meant to facilitate use and experimentation with AI systems at Yale.
