Dubai, United Arab Emirates (CNN) — Researchers in Denmark say they have used powerful machine learning algorithms to predict certain aspects of human life, including how soon someone will die.

Their study, published in the journal Nature Computational Science, shows how a machine learning algorithm model called life2vec can predict a person's life outcomes and actions when given very specific data.

“With this data, we can make any kind of prediction,” said Sonny Lehmann, lead author of the study and a professor at the Technical University of Denmark. However, the researchers note that it is a “research prototype” and cannot perform “real-world tasks” in its current state.

Lehmann and his colleagues used data from a national registry in Denmark describing a diverse group of 6 million people. It covers information related to key aspects of life such as education, health, income and occupation from 2008 to 2016.

The researchers adapted language processing techniques and developed a vocabulary for life events so that the “life2vec” model could interpret sentences based on the data.

The algorithm then learned from that data, Lehmann explains, and was able to make predictions about certain aspects of people's lives, including how they might think, feel, act, and even whether someone would die in the next few years.

The team used data from January 1, 2008 to December 31, 2015 for a group of more than 2.3 million people between the ages of 35 and 65 to predict how soon someone would die.

Lehmann points out that this group was chosen because it is difficult to predict deaths in this age group.

The Life2vec model used the data to estimate the probability that a person would survive four years after 2016.

“To test how well the life2vec model did, we selected a group of 100,000 people, half of whom survived and the other half died,” Lehmann said.

Unlike the algorithm, the researchers were aware of those who died after 2016. To test that, they had an algorithm that made individual predictions about whether or not someone lived past 2016. The results were impressive: the algorithm was correct 78% of the time.

The report asserted that the Life2vec model outperformed other recent models and benchmarks by at least 11% in predicting mortality outcomes more accurately.

The researchers found that men were more likely to die after 2016 than skilled workers, such as engineers, or those with mental health problems such as depression or anxiety, which also lead to early death.

At the same time, holding a managerial position or having a high income often pushes people towards the “survival” column.

However, the research had several limitations, as the report notes that “the trials were not randomized, and the researchers did not intentionally individualize the results during the trials.”

The researchers only looked at data over an eight-year period, and although everyone in Denmark appears in the national register there may be sociodemographic biases in the sample.

The researchers also noted that the study was conducted in a wealthy country with strong infrastructure and a strong health system. It is unclear whether Life2vec's findings would apply to other countries, such as the United States, based on economic and social disparities.

Lehman acknowledges that the algorithm is “monstrous and crazy, but it's actually done a lot of work, especially by insurance companies.”

Dr. Arthur Caplan, chair of the Department of Clinical Ethics at New York University's Crossman School of Medicine, agrees that insurers will be eager to outbid consumers as models like Life2vec become more commercial.

He added: “It makes selling insurance very difficult. You can't insure risk if everyone knows what the risks are.”

However, Kaplan, who was not involved in the new research, points out that the Life2Week model does not predict at what age a person will die or how they will die. For example, an algorithm cannot predict whether a person will be killed in a car accident.

Kaplan expects more advanced predictive models to emerge within five years.

“We have great companies with huge databases that will make recommendations about what to do to extend your life,” he said.

Ultimately, Kaplan says, “Using AI to predict when we might die removes the one aspect of our lives that keeps them interesting: the mystery.”

Kaplan expects more advanced predictive models to emerge within five years.

“We have great companies with huge databases that will give you advice on what to do to extend your life,” he explained.

He added: “We're worried about robots taking over the world and deciding we don't need them anymore. But what we should be worried about is the ability of robots to manipulate information and predict our behavior so we get a life. It's so predictable that it has some value. takes away from us.”