Dr. Jack Peter Gallifant is an accomplished researcher bridging artificial intelligence and clinical medicine with expertise in developing robust and equitable healthcare AI systems. Currently serving as a Postdoctoral Research Fellow in the AI in Medicine program at Harvard Medical School and Mass General Brigham he brings unique perspective from his clinical background as a former NHS physician. Dr. Gallifant earned his medical degree from the University of East Anglia and completed a Master of Science in Human and Applied Physiology at King's College London establishing a solid foundation for his interdisciplinary work at the intersection of healthcare and technology. His career trajectory from clinical practice to AI research leadership exemplifies his commitment to translating technological advances into meaningful healthcare improvements while addressing systemic health disparities.
Dr. Gallifant's pioneering work focuses on developing frameworks for evaluating AI systems in healthcare settings with particular emphasis on interpretability robustness and identifying bias in algorithmic decision-making. He led the development of TRIPOD-LLM a comprehensive framework for transparent reporting of large language models in healthcare applications which was published in Nature Medicine following expert consensus building. His systematic analysis of sparse autoencoders for interpretable feature extraction from large language models represents a significant methodological contribution to understanding black box AI decision processes. Dr. Gallifant's research on disparity dashboards for continuous evaluation of AI models across different demographic subgroups has been instrumental in advancing the field of equitable AI deployment in clinical settings.
Beyond his technical contributions Dr. Gallifant has established himself as a thought leader in promoting responsible AI development through his active participation in professional organizations and educational initiatives. He is a key contributor to MIT Critical Data where he teaches physicians and data scientists how to evaluate clinical data and develop algorithms that work equitably for diverse populations. His recent peer review of the GPT-4 technical report published in PLOS Digital Health has shaped important conversations about transparency in large language model development. Currently focusing on developing standardized benchmarks for evaluating multimodal AI systems in healthcare across international contexts Dr. Gallifant continues to lead critical research that will ensure AI technologies enhance rather than exacerbate healthcare disparities globally.
