A “death calculator”: this AI is capable of predicting premature deaths of 35-65 year olds almost every time

A “death calculator”: this AI is capable of predicting premature deaths of 35-65 year olds almost every time


A “death calculator” could soon see the light of day. Danish researchers have been working for several years on an algorithm capable of predicting the major stages of life… until death. The results of the tests, rather conclusive, were published in the journal Nature Computational Science.

To develop Life2vec, scientists used an operating model similar to that of ChatGPT. But instead of processing textual data, the algorithm analyzes life stages such as birth, studies, social benefits or even working hours.

“From a certain point of view, life is just a series of events: people are born, go to the pediatrician, go to school, move, get married, etc.,” the study explains. Here we exploit this similarity to adapt innovations in natural language processing to examine the evolution and predictability of human lives based on detailed event sequences. »

78% success in predicting death

The researchers used anonymized data from nearly six million Danes. “With a very young cohort of people aged between 35 and 65, we try to predict, based on an eight-year period (2008 to 2016), whether the person will die in the next four years, until 2020. The model does this very well, better than any other algorithm,” explains Sune Lehmann, professor at Danamerk Technical University (DTU).

This young age group was not chosen at random. Deaths, usually few in number, make it possible to verify the reliability of the program, according to the researchers. On death, the algorithm is right in 78% of cases, on migrations, in 73%.

However, the tool is not yet ready for use by the general public. It still has many biases. “For the moment, it is a research project which explores the field of possibilities (…), we do not know if it treats everyone equally,” explains the researchers.

Fertility, obesity, cancer…

Besides death, the algorithm could be used more widely to predict fertility, obesityas well as the risk of cancer. “It could also predict whether you are going to make a lot of money,” says Professor Lehmann.

For the academic, the project presents a scientific counterweight to the algorithms developed by Gafam (Google, Apple, Facebook, Amazon, Microsoft). “They can also build models like this, but they don’t make them public, they don’t talk about them,” he says. And added: “We can hope that they develop them only to make us buy more products. »



Source link