Advancements in AI: The ‘Doom Calculator’ Predicts Lifespan with Remarkable Accuracy
Are you curious about knowing the exact time of your death? Science might soon offer you this choice.
The latest development in this field is an AI algorithm, nicknamed ‘the doom calculator’ by the U.K.’s Daily Mail. This tool has successfully predicted the death of individuals within a four-year period in over 75% of cases.
The team behind this project includes researchers from Denmark and the U.S. Their findings were recently published in Nature Computational Science. The AI, similar in concept to ChatGPT but non-interactive, was named life2vec. It analyzed a vast range of data – from age and health to job and income – of more than 6 million Danes, provided by the Danish government, which supported the study.
This model learned to understand life narratives, such as “Francisco received a sum of money for his job at a castle in 2012” or “Hermione chose elective classes during her third year at boarding school,” as detailed in the research paper.
As it developed, life2vec began to construct comprehensive life stories for individuals. According to Sune Lehmann, a professor at the Technical University of Denmark and former postdoctoral fellow at Northeastern, a human life can be seen as a long series of events. His colleague, Tina Eliassi-Rad, a computer science professor at Northeastern University in Boston, also collaborated on this research.
Remarkably, the AI was accurate in predicting deaths by 2020 in about 78% of cases.
However, participants in the study weren’t informed of their death predictions. Lehmann emphasized the irresponsibility of doing so in an interview with the New York Post.
The research identified certain factors linked to earlier deaths, such as mental health issues, being male, or having certain professions. Conversely, leadership roles and higher incomes tended to correlate with longer life spans.
Lehmann also mentioned the AI’s ability to predict personalities and decisions, like moving abroad.
In terms of public access to this ‘doom calculator,’ it’s not available yet due to privacy concerns. Lehmann and his team are exploring ways to share their findings more broadly while ensuring the privacy of the individuals involved.
