December 21, 2023: In an age where technology increasingly intersects with all aspects of life, a groundbreaking AI death calculator developed by researchers in Denmark and the United States has emerged, capable of predicting the lifespan of individuals with a striking 78% accuracy.
This AI tool, dubbed “life2vec,” utilizes a blend of personal details such as income, profession, residence, and health history to forecast life expectancy.
The brainchild of Sune Lehmann, a professor at the Technical University of Denmark, and his team, life2vec employs a technology akin to that behind ChatGPT, analyzing human lives as sequences of events.
Lehmann explains, “Just like words in sentences, events follow each other in human lives.”
This similarity allows the AI to closely examine a person’s past to compute life outcomes.
In their study, Lehmann’s team scrutinized a diverse group of 6 million Danish people from 2008 to 2020.
The AI was tasked with determining which individuals would live at least four years beyond January 1, 2016.
Researchers input detailed data into the algorithm, represented by specific digital tokens for events like a forearm fracture or working in a tobacco shop.
The AI’s predictions were not just limited to life expectancy. Lehmann notes, “This model can predict almost anything,” including personality traits and decisions to move internationally.
The choice to predict death was driven by its longstanding research interest, particularly among insurance companies, providing a benchmark for what was possible.
Despite its impressive predictive power, Lehmann emphasizes the importance of responsible use. The algorithm’s predictions are not intended for making judgments in scenarios like insurance policies or employment.
Instead, the focus is on understanding the possibilities and limitations of such predictions.
As for public access, life2vec is not yet available, and Lehmann assures that if it ever becomes mainstream, it won’t be used to inform individuals of their death predictions, as that would be “very irresponsible.”
The study sheds light on various factors influencing lifespan, with being male, having a mental health diagnosis, or working in a skilled profession linked to earlier death.
Conversely, higher income and leadership roles correlate with a longer life. While the AI won’t be used to provide personal death predictions, it opens a window into understanding factors that could contribute to longevity.