Is your death near or do you have a long life left to live? These kinds of questions time and again occupy our minds and we remain curious about our life span.
But now, with the advancement of artificial intelligence technology, researchers are also trying to solve all our curiosities. A new AI death calculator has been developed which is said to predict when a person will die with a scary accuracy of 78 per cent.
“We use the technology behind ChatGPT (something called transformer models) to analyze human lives by representing each person as the sequence of events that happens in their life,” said Sune Lehmann, lead author of the study “Using a sequence of life-events to predict human lives”, which was published in December 2023, reported the New York Post.
Lehmann, who is a professor of network and complex systems from the Technical University of Denmark along with the co-authors, in the report introduced the world to an algorithm called as “life2vec”, which has been using selective information from the life of an individual – which includes health history, residence, income and profession – to find their life expectancy, which they claim to be 78 per cent accurate.
“We use the fact that in a certain sense, human lives share a similarity with language,” said Lehmann. “Just like words follow each other in sentences, events follow each other in human lives,” the professor added.
The newly-built model life2vec is a little different from the famous ChatGPT, which is a bot that is being employed by tech wizards to help them find their dream jobs and even find the perfect outfit. This AI model is capable of computing the outcomes of life by closely examining a person’s past.
“This model can predict almost anything,” said Lehmann, while further emphasising that his research team has used specialised programmes to foretell the personalities of people.
“We predicted death because it’s something people have worked on for many years (for example, insurance companies), so we had a good sense of what was possible,” he added.
The troop of Lehmann examined a heterogeneous subject population of around 6 million Danish people, who were of different sex and age, between 2008 and 2020. The life2vec was used by the analysts to discover which of the subjects would likely continue living for at least four years after January 1, 2016.
“The scale of our dataset allows us to construct sequence-level representations of individual human life trajectories, which detail how each person moves through time,” the report read. “We can observe how individual lives evolve in a space of diverse event types (information about a heart attack is mixed with salary increases or information about moving from an urban to a rural area),” it added.
The AI-specific information, regarding each study participant, was fed by the researchers using simple language like, “In September 2012, Francisco received 20,000 Danish kroner as a guard at a castle in Elsinore” or “During her third year at secondary boarding school, Hermione followed five elective classes.”
Comments