Validation requirements for AI-based intervention-evaluation in aging and longevity research and practice
Georg Fuellen, Anton Kulaga, Sebastian Lobentanzer, Maximilian Unfried, Roberto A. Avelar, Daniel H. Palmer, Brian K. Kennedy
Abstract
The field of aging and longevity research is overwhelmed by vast amounts of data, calling for the use of Artificial Intelligence (AI), including Large Language Models (LLMs), for the evaluation of geroprotective interventions. Such evaluations should be correct, useful, comprehensive, explainable, and they should consider causality, interdisciplinarity, adherence to standards, longitudinal data and known aging biology. In particular, comprehensive analyses should go beyond comparing data based on canonical biomedical databases, suggesting the use of AI to interpret changes in biomarkers and outcomes. Our requirements motivate the use of LLMs with Knowledge Graphs and dedicated workflows employing, e.g., Retrieval-Augmented Generation. While naive trust in the responses of AI tools can cause harm, adding our requirements to LLM queries can improve response quality, calling for benchmarking efforts and justifying the informed use of LLMs for advice on longevity interventions. • We list 8 Requirements for good recommendations by Artificial Intelligence (Large Language Models) in giving longevity-related recommendations. • 3 Requirements are of general importance, including correctness of the recommendations. • 3 Requirements are more specific to recommendations about interventions, including an emphasis on causal evidence. • 2 Requirements are specific to longevity-related interventions, these are consideration of longitudinal data, and of evidence related to known mechanisms of aging. • Adding a list of our Requirements to the prompt improves the quality of the recommendations made Large Language Models.