Scientists have developed a revolutionary chatbot, called “Life2vec”, capable of estimating a person’s life expectancy with a remarkable 78% accuracy.
Artificial intelligence amazes us every day, but also terrifies us. PHOTO Archive
This artificial intelligence model, the result of a collaboration between Danish and American researchers, analyzes a vast database containing demographic and medical information for millions of individuals.
Life2vec can also determine how much money you’ll have when you die, according to metro.co.uk
Scientists fed data from Danish demographic and health records for six million people into the model. Like ChatGPT, it has learned to predict what will come next based on requirements.
But unlike ChatGPT, this AI uses information such as income, occupation and medical records to determine how long a person will live.
Previous research has already established that certain life factors can lead to longer or shorter lives. For example, being male, a smoker, or a diagnosis of poor mental health can all contribute to a shorter lifespan. Higher income and having a leadership role can lead to a longer life.
Each of these factors has a code in the Danish dataset, such as “S52” for a broken forearm or “IND4726” for working in a tobacco shop, which the team turned into words.
This allowed people’s lives to be summarized in data-rich sentences, including things like profession, income, injuries and pregnancy history.
“In September 2012, Francisco received twenty thousand Danish kroner as a guard at a castle in Elsinore” and “During her third year at boarding school, Hermione took five electives” are examples of stories written about topics.
The AI then fits all these words together like a puzzle to determine how long that person will live.
The team tested Life2vec on a group of people aged 35 to 65, half of whom died between 2016 and 2020.
Life2vec predicted who would die and who would live with astonishing accuracy – 78%.
It was also able to predict personality test scores more accurately than models trained specifically for the task.
The chatbot is not yet available for public use.
The team, led by Professor Sune Lehmann Jorgensen from the Technical University of Denmark, noted that because the model was trained exclusively on data from Denmark, the results may not be as accurate for people in other countries.
They also stressed that AI should not fall into the hands of big business. “Clearly, our model should not be used by an insurance company”, said Professor Jorgensen.
However, he added that similar technology already exists.
“They are probably already being used for us by big tech companies who have tons of data about us and use it to make predictions about ushe said.
Writing in the journal Nature Computational Science, the team adds: “Our framework allows researchers to uncover potential mechanisms that impact life outcomes, as well as the associated possibilities for personalized interventions.”
