Danish researchers have developed an algorithm to predict when you'll die

A group of researchers from a Danish university has developed a model called the “Death Calculator”, an algorithm for predicting the stages of life until the end of life that seeks to show the risks of commercial use of this data.

“It's a very general framework that makes it easy to predict human life,” he told AFP. “You can predict anything as long as you have data.” Sonny Lymana professor at the Technical University of Denmark (DTU), and one of the authors of the study published in the journal Nature Computational Science.

According to him, the possibilities are endless. “It can predict health outcomes. It can predict fertility or obesity, or maybe who's going to get cancer or not. But it can also predict whether you're going to make a lot of money,” he added.

specific, life2vec It uses a similar model to ChatGPT But instead of dealing with textual data, it analyzes statistics such as birth, education, social benefits or 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 from one house to another, and get married,” the study said. “We exploit this similarity here to adapt automatic natural language processing innovations to examining the evolution and predictability of human life based on detailed sequences of events,” he says.

Millions of people have been analysed

This report is based on anonymous data from millions of Danes, collected by the Nordic country's National Statistical Institute. Sequence analysis makes it easy to predict what will happen until the end. about death, The algorithm is correct in 78% of casesAnd on migrations by 73%.

“With a sample of people aged 35 to 65, we seek to predict based on an eight-year period, from 2008 to 2016, whether a person will die in the next four years, up to 2020. The model does this very well. . “Better than any other algorithm,” says Lehman, who doesn’t use his formula in personal situations.

This age group, in which deaths are usually low, makes it easy, according to researchers, to verify the reliability of the program.

But the tool is not ready for use by the general public. “Right now it is Research project Who explores the field of possibilities (…), and we do not know if he treats everyone the same way!”, he explains.

The long-term impact, social relationships, and predictability of the life course still need to be studied.

Scientific weight

For an undergraduate, the project represents a scientific counterweight to algorithms developed by giants such as Google, Apple, Facebook, Amazon and Microsoft.

“They can also build models of the same type, but they don't advertise or talk about it,” he says. “We can assume they perfected it just to make us buy more products,” he added.

For him, it is important to have a public and open counterweight to begin to understand what can be done with data of this kind.

And much more so because algorithms of this kind are certainly already being used in the insurance industry, as the ethics expert points out, Pernell Tranberg.

“They definitely put us into groups (…) and that can be used against us because they can force us to pay higher insurance, or refuse us a loan from the bank or public medical care because one of us will die anyway.” “He adds.

He confirms that “there are no cases of personal data leakage” with the National Institute of Statistics, and “the data is not individual.” However, with the development of artificial intelligence, “everything is accelerating.”