While generative AI holds promise in facilitating access to current medical information and supporting clinical decision-making, ensuring its safe, responsible, and ethical application in clinical settings presents challenges. Join this session to learn about the evaluation methodology used to evaluate the performance of Elsevier’s ClinicalKey AI. We will share insights from Elsevier’s evaluation team’s experience establishing and implementing a framework for assessing the correctness and potential harmfulness of responses generated by the conversational search engine embedded within the tool. Presented by Elsevier