Researchers have developed an algorithm that uses the story of a person's life to predict their future and it is accurate about 78 per cent of the time.
They now have a model that understands everyday language and can guess at things like if someone might die young or how much they could earn in their lifetime.
The study shows being a boy, having mental health problems or working skilled jobs can mean shorter lifespans, while earning lots of money or leading other people can add years to your life. The system, life2vec, treats each part of your life story like bits in a sentence to guess what's coming up based on whats already happened.
Scientists have developed a new AI system called life2vec that uses data from 2008 to 2016 to predict simple questions like "death within four years?" for any given person. Scarily enough, it correctly guessed who would die by 2020 over three-quarters of the time.
Sune Lehmann, the lead researcher, says that the system isn't open to the public or companies, to meet GDPR standards and keep private information safe: "We are actively working on ways to share some of the results more openly, but this requires further research to be done in a way that can guarantee the privacy of the people in the study."
The report was published in Nature Computational Science. Sune Lehmann, a professor at Denmark Technical University, explained how life2vec works. He said it's similar to language models like ChatGPT: it learns from a vast amount of pre-existing written data, except life2vec is taught using data from people's lives.
Sentences like "In September 2012, Francisco received twenty thousand Danish kroner as a guard at a castle in Elsinore" or "During her third year at secondary boarding school, Hermione followed five elective classes." were given different tokens by the scientists. These pieces of data were all mapped out in relation to each other.
@ISIDEWITH11mos11MO
@ISIDEWITH11mos11MO
