Advancements in artificial intelligence have taken a fascinating turn as researchers from Denmark and the U.S. introduce “the doom calculator,” an AI algorithm that can predict whether individuals will die within four years, achieving a remarkable accuracy rate of over 75%.
Published in the Nature Computational Science online journal, the project utilized an AI machine-learning transformer model named “life2vec,” comparable to ChatGPT but without interactive features. Researchers, including Sune Lehmann, a professor of networks and complexity science at the Technical University of Denmark, and Tina Eliassi-Rad, a professor of computer science at a university in Boston, delved into the vast dataset provided by the Danish government, encompassing data on more than 6 million individuals.
Life2vec was trained to process various life-related data, including age, health, education, employment, income, and other significant life events. The model, evolving over time, learned to assimilate information about individuals’ lives by interpreting sentences like “In September 2012, Francisco received 20,000 Danish kroner as a guard at a castle in Elsinore” or “During her third year at secondary boarding school, Hermione followed five elective classes,” creating “individual human life trajectories.”
Sune Lehmann described the evolving model’s capability to construct “the whole story of a human life” as a giant, lengthy sentence encapsulating the myriad experiences one undergoes. While the project sparks intrigue in the realm of AI, it also raises ethical considerations regarding the potential implications of predicting human mortality. The researchers’ findings open avenues for further exploration into the evolving landscape of AI applications in understanding and analyzing the complexities of human existence.
