Death Calculator: An Artificial Intelligence Tool Predicts Your Date of Death
Researchers in Denmark have developed an artificial intelligence tool they’ve dubbed the “Death Calculator,” using data from millions of individuals.
The aim is to help predict an individual’s life stages until the end, in a move intended to raise awareness about the power and risks of this technology.
A General Framework for Predicting Human Life
One of the authors of the study published in the journal “National Computational Science,” Professor at the Technical University of Denmark, Sonny Lehmann, explained to “France Presse” that the tool is “a very broad general framework for predicting human life, capable of predicting anything if it has training data.”
He emphasized that the possibilities are open and endless, with the tool “capable of predicting health outcomes, anticipating fertility or obesity, or perhaps whether someone will get cancer or not, and it can even predict if a person will earn a lot of money.”
Practically, “Life2Vec” uses a similar operational model to the “ChatGPT” model, but instead of processing textual data, it analyzes life stages such as birth, education, or even working hours.
The study noted that “life is essentially a series of events: people are born, they go to the pediatrician, they go to school, they move from one home to another, they get married, etc.”
And it added, “We here benefit from this similarity to adapt innovations in natural language processing to the requirements of studying human life evolution and prediction, based on a detailed sequence of events.”
Data from 6 Million Danes
The tool is based on anonymized data from about six million Danes collected by the National Statistics Institute.
The sequence analysis allows the prediction of remaining stages until the end of life. Regarding mortality, the algorithm achieves a success rate of 78%, and in cases of migration, a rate of 73%.
Lehmann said, “With a very small group of people aged 35 to 65, we’re trying to predict, based on eight years of data from 2008 to 2016, whether the person will die in the next four years, until 2020, and the model does this very well, better than any other algorithm.”
This age group, where deaths are usually rare, allows for verifying the reliability of the program, according to the researchers.
But the tool is not yet ready for use by the general public as it still has flaws, and it is “currently just a research project exploring possibilities, and we don’t know if it will treat everyone fairly.”
Scientific Balance
The project, according to the professor, represents a scientific balance compared to the algorithms developed by major technology companies known as “Gafam,” namely “Google,” “Apple,” “Facebook,” “Amazon,” and “Microsoft.”
He said, “These companies can also build models like this tool, but they don’t announce them or talk about them.” The researcher noted that these companies simply use these algorithms to encourage the public to “buy more products.”
He stressed the importance of “having a balanced, available, and open weight that shows what can be done with data of this kind.”
Data Ethics Expert Pernille Tranberg pointed out that such algorithms are already used in the insurance field.
She added, “We have certainly been classified into groups, and this can be used against us, to the point where we have been pushed, for example, to buy insurance at a higher price, or to prevent us from getting a bank loan or government healthcare because we will die anyway.”
However, this does not apply to the research project, as it is not intended for individual use, as the identity of its sources remains confidential.
She assured that “there have been no examples of personal data leaks” recorded with the National Statistics Institute, and that “the data is not named,” but she cautioned that “everything is accelerating” due to the development of artificial intelligence.
Commenting on some attempting to exploit the idea for commercial purposes, Tranberg said, “On the internet, we already see age predictions, indicating the age at which the person will arrive, and some of them are not credible.”
