Speech given at Int. Conference on Artificial Intelligence: Risks, Regulations and Roadmaps, Dec 7th 2023. The conference addresses the balancing act of regulation: seeking to ensure neither too little nor too much. While we'll present and discuss major societal threats posed by AI, we'll also consider the missed opportunities if stringent regulations hinder AI-based scientific, economic, and general human progress.
Short abstract
For AI to deliver its promise, we need to invest in the "who" and the "how", and not just the "what". We need to fix the medical knowledge system that feeds large language models, and this requires diversifying who are sitting at the table (funders and professional societies) and who are preparing the food (researchers).
The biggest threat to AI is the lack of societal understanding of what it can and cannot do, and the risks associated with its use. The expectation that AI can "recode" the world without dismantling the legacy systems that have preserved the age old problems that we face is misguided.
There is an opportunity to inch us closer to a better care delivery system, but it will require us to design an equity-focused sociotechnical ecosystem. Engineering AI in a vacuum that does not build capacity nor include cultural transformation of how we learn and how we work together will be a waste of time and money, and a huge opportunity cost.
Bio
Leo Anthony Celi is currently the Clinical Research Director and Senior Research Scientist at the Laboratory for Computational Physiology at MIT and a practicing intensivist at the Beth Israel Deaconess Medical Center in Boston, Dr. Celi’s work focuses on scaling clinical research to be more inclusive through open access data and software, particularly for limited resource settings; identifying bias in the data to prevent them from being encrypted in models and algorithms; and redesigning research using the principles of team science and the hive learning strategy.
Dr. Celi completed his medical degree at the University of the Philippines, followed by postgraduate training in internal medicine, critical care medicine, infectious diseases and biomedical informatics at Cleveland Clinic, Harvard, Stanford and MIT. He has published numerous papers in machine learning not just in critical care medicine but across different specialties such as ophthalmology, radiology, surgery, nursing, pharmacy, bioethics, among others.
