As artificial intelligence (AI) tools become increasingly integrated into academic research workflows, librarians and student researchers are exploring systematic methods for evaluating their effectiveness. Building on previously published frameworks, this study introduces a rubric-based approach to assessing AI tools across three key areas: information discovery, search, and reviews. Notably, Undermind and the paid version of Elicit emerged as top performers. The study highlights the importance of nuanced evaluation methods and suggests that paid versions of AI tools may offer enhanced functionality for academic research.
Patterson et al. (Sun,) studied this question.