Monday, March 4, 2024

AI’s Latest Feat: Death Calculator Claims Startling 78% Accuracy in Predicting Your Demise

Share

Questions about the duration of one’s life often linger in our minds, prompting curiosity about our life expectancy. The recent strides in artificial intelligence (AI) technology aim to address these curiosities, introducing a new AI death calculator purported to predict an individual’s death date with an unsettling accuracy rate of 78%.

Lead author Sune Lehmann, a professor of network and complex systems at the Technical University of Denmark, detailed the underlying technology behind the AI death calculator in a study titled “Using a sequence of life-events to predict human lives,” published in December 2023. The technology, named “life2vec,” leverages the principles of transformer models, the same technology found in ChatGPT, to analyze human lives by representing each person as a sequence of life events.

Lehmann explained that the life2vec algorithm utilizes selective information from an individual’s life, encompassing factors such as health history, residence, income, and profession, to estimate life expectancy with a claimed accuracy of 78%.

“We use the fact that, in a certain sense, human lives share a similarity with language. Just like words follow each other in sentences, events follow each other in human lives,” stated Lehmann.

Distinguishing itself from the widely known ChatGPT, which assists users in job searches and fashion choices, life2vec focuses on predicting outcomes by closely examining an individual’s past experiences.

According to Lehmann, the model can predict a wide array of information, with the research team using specialized programs to forecast individuals’ personalities. Lehmann emphasized that death prediction was chosen due to the extensive history of research in the field, particularly by insurance companies.

The research team, led by Lehmann, conducted their study on a diverse subject population of approximately 6 million Danish individuals, varying in sex and age, covering the period from 2008 to 2020. Utilizing life2vec, the analysts aimed to identify which subjects were likely to survive for at least four years beyond 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. 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),” the report highlighted.

Researchers input AI-specific information for each study participant using straightforward language, such as “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.”

Read more

Recommended For You