New AI algorithm makes predictions about both life and death

A team of researchers from the Technical University of Denmark, Northeastern University and the University of Copenhagen, among others, is using artificial intelligence (AI) and data from millions of people in Denmark to help make predictions about the different stages of an individual’s life — and even death.

The creators of the life2vec program set out to examine patterns and relationships that deep-learning programs uncover, thereby enabling the program to make predictions about a range of health or social “life-events.”

Wherever there is training data, the researchers suggested, there is an opportunity for making predictions about human lives.

“It could predict health outcomes. So it could predict fertility or obesity, or you could maybe predict who will get cancer or who doesn’t get cancer. But it could also predict if you’re going to make a lot of money,” the researchers explained.

Using an algorithm that employs similar processes as those used by ChatGPT, the technology examines a wide range of variables that impact life — for instance, birth, education, social benefits and even work schedules. Using this data, the team sought to “examine the evolution and predictability of human lives based on detailed event sequences.”

Data about such sequences of events — including details about birth, visits to the pediatrician, starting school, moving to a new locations, getting married — comes from the anonymized data of roughly six million Danes, as gathered by the official Statistics Denmark agency.

The researchers suggest that analyzing such sequences of events makes it possible to predict life outcomes right up until the very end of life. For instance, the algorithm correctly predicted death in 78% of cases and moves to another city or country in 73% of cases using this data.

Although this algorithm has reportedly outperformed other so-called “death calculator” algorithms, the researchers caution that it is not yet ready for use beyond a research setting. Further, the team insists that the software is currently private and unavailable on the internet.

The team details their research in the article, “Using sequences of life-events to predict human lives,” which appears in the journal Nature Computational Science.

UNLIMITED FREE ACCESS TO THE WORLD’S BEST IDEAS

SIGN UP TO SEE MORE

This is embarrasing…

An error occurred while processing the form. Please try again in a few minutes.

Customize Your GlobalSpec Experience

This is embarrasing…

An error occurred while processing the form. Please try again in a few minutes.