AI Meets Evidence: Practical Applications for Librarians in the Review Workflow
Sponsored by Sysrev
Recorded on 11/18/2025Posted in Library Technology and IT, Scholarly Communication and Research
Learn how to integrate AI into evidence synthesis and help researchers utilize AI tools more efficiently.
Access the presentation slides.
Summary:
Evidence synthesis refers to a suite of methods used to systematically search for, review, and integrate findings from research to answer targeted questions, identify knowledge gaps, and inform practice and policy. With the rapid growth of published research, there is an urgent need to make this process more efficient—without compromising quality or rigor. Artificial intelligence (AI), particularly generative AI and large language models, offers both new opportunities and unique challenges in the context of evidence synthesis. As librarians, we are increasingly asked to support students and researchers in navigating this evolving landscape, helping them choose appropriate tools, methods, and workflows aligned with their review objectives.
This webinar will provide a practical overview of current approaches to integrating AI into evidence synthesis, with a focus on the study selection and data extraction stages. We will explore findings from recent research on what works (and what doesn’t), share techniques for effective prompt development, and discuss key considerations around software selection, workflow design, and transparent reporting. Participants will come away with a clearer understanding of the role AI can play in evidence synthesis and better equipped to guide researchers in using AI tools effectively, efficiently, and responsibly.
Speakers:
Sarah Young
Social Sciences Librarian and Director, Evidence Synthesis ProgramCarnegie Mellon UniversitySarah Young is a social sciences librarian and Director of the Evidence Synthesis Program at Carnegie Mellon University, as well as a Research & Strategy Consultant for Sysrev. In her role as a liaison librarian, Sarah provides teaching, research and collections support to public policy, information systems and management programs. As an evidence synthesis specialist, Sarah provides methodological expertise in search, retrieval and information management for evidence synthesis across disciplines, with a focus on social sciences. She serves as co-convener of the Campbell Collaboration’s Information Retrieval Methods Group and is an Associate Editor for Research Synthesis Methods. Sarah has published research on evidence synthesis methods and has developed and conducted training for librarians and researchers globally. She holds graduate degrees in library and information science and international development and a graduate certificate in program evaluation.
TJ Bozada
Product LeadSysrevTJ Bozada is the Product Lead of Sysrev, a web-based platform that supports literature reviews, evidence synthesis, study selection, and data extraction. In his role, TJ oversees all non-engineering aspects of the platform, from product roadmap and innovation to direct collaboration with research teams. He works closely with researchers at every stage of the review process, providing training on Sysrev’s tools, designing data extraction workflows, refining large language model prompts, and developing strategies that balance GenAI capabilities with human expertise in screening and extraction. His work emphasizes transparency, reproducibility, and the responsible integration of AI into scholarly research. TJ holds dual Bachelor of Science degrees in Biomedical Engineering and Applied Mathematics & Statistics from Johns Hopkins University.
All Choice webinars offer Zoom’s auto-generated AI captioning. If you have any questions, please contact: scofer@ala.org.
This webinar will be recorded. All registrants should receive a follow-up email with a link to the recording. Webinar recordings are also available on Choice’s YouTube channel.
This webinar is a paid sponsorship opportunity. The products, services and opinions presented herein do not constitute a CHOICE, ACRL, or ALA endorsement of any kind.
