Researchers have developed a new artificial intelligence (AI) system named life2vec, which treats human lives as language and accurately predicts various life details, including mortality, international moves, and personality traits. The Danish researchers behind the study utilized data from millions of residents, such as birth dates, sex, employment, location, and universal healthcare system usage, to train the machine-learning model. Over a four-year period, life2vec demonstrated over 78% accuracy in predicting mortality, surpassing other predictive methods. The model’s flexible architecture allows for easy adjustments, making it a promising tool for predicting various aspects of human life.
Life2vec, created by the researchers, introduces a novel approach to predicting and analyzing the trajectories of people’s lives. The model processes individual data into unique timelines of events, such as salary changes and hospitalizations, resembling a language processing tool. Beyond mortality predictions, life2vec showed early promise in connecting personality traits with life events, achieving a 73% accuracy rate in predicting people’s moves out of Denmark and self-reported responses to personality questionnaires.
Also Read: IMF warns that Artificial Intelligence Will Affect 40% of Office Jobs
Matthew Salganik, a Princeton University professor of sociology specializing in computational social science, notes that life2vec’s developers employ a unique style not previously used in similar studies. The model’s adaptable architecture enables easy adjustments and fine-tuning, making it a versatile tool for predicting various unexplored aspects of human life. Medical professionals have expressed interest in developing health-related versions of life2vec to explore population-level risk factors for rare diseases.
The life2vec model opens up possibilities for uncovering hidden societal biases and understanding unexpected connections between factors like professional advancement, age, or country of origin. The AI tool has potential applications in exploring the impact of relationships on quality of life and salary, shedding light on previously unrecognized societal biases.
