A groundbreaking study published in the journal Nature Computational Science introduces Life2Vec, a novel deep-learning program designed to forecast various life events ranging from health outcomes to financial success. Spearheaded by Sune Lehmann, a professor at the Technical University of Denmark (DTU), the research team aims to explore the potential predictive capabilities of the algorithm by analyzing intricate patterns and relationships in human life sequences.
“It’s a very general framework for making predictions about human lives. It can predict anything where you have training data,” Lehmann explained to AFP.
With an extensive dataset comprising anonymized information from approximately six million Danes, sourced from Statistics Denmark, the algorithm delves into variables such as birth, education, social benefits, and work schedules to anticipate life trajectories. Lehmann emphasizes the algorithm’s versatility, stating, “It could predict health outcomes… but it could also predict if you’re going to make a lot of money.”
Despite early speculations labeling the program as a “death calculator,” Lehmann assures that Life2Vec remains a research project and is not publicly available. He stresses the importance of open dialogue surrounding such advancements, particularly in light of potential ethical concerns regarding data usage and privacy.
“We look at early mortality… The model can do that really well, better than any other algorithm that we could find,” Lehmann said, underscoring the algorithm’s impressive accuracy in predicting death within a specific age bracket.
While the research holds promise for understanding life’s predictability, experts caution against commercializing such algorithms prematurely. Pernille Tranberg, a Danish data ethics expert, warns of potential discriminatory practices by businesses utilizing similar predictive models.
“They probably put you into groups and say: ‘Okay, you have a chronic disease, the risk is this and this.’ It can be used against us to discriminate us…,” Tranberg explained.
