Scientists from Denmark and the US have developed an algorithm that uses a person’s life history to predict how they will live and when they will die.

According to a new study, the model called “life2vec”, it is accurate about 78% of the time, which puts it on par with other algorithms designed to predict similar life outcomes.

But unlike other models, works like a chatbot, using existing details to predict what will come next.

It was built by scientists in Denmark and the US who used a machine learning algorithm with data from more than 6 million people, feeding it all kinds of information like: income, occupation, place of residence, injuries, pregnancy, etc.

Their final result was a model that can process plain language and generate predictions about a person’s likelihood of dying prematurely or to increase his income during his lifetime.

Some of the factors that can lead to early death include: being male, having a diagnosed mental illness or being in a skilled profession. Things associated with longer lifespans include; higher income or leadership role.

How does it work

Just like ChatGPT users they ask him to write a song, poem or essay, scientists can ask life2vec simple questions like “death within four years?” for a specific person.

The model trained on data from 2008 to 2016.

Based on their population data, correctly predicted who had died by 2020 more than three-quarters of the time.

Similar technologies for predicting life events and human behavior are already being used today by technology companies that, for example, they monitor our behavior on social networks, present us with highly accurate profiles and use these profiles to predict our behavior so as to affect us.

*The research was published in Nature Computational Science.