'Scarily accurate' AI calculator can predict when you're going to die

vt-author-image

By James Kay

Article saved!Article saved!

A new AI calculator can apparently predict when you're going to die, and some have described it as "scarily accurate".

Death is something that some people aren't really bothered by and they've already accepted that it'll happen and just get on with their life in the meantime.

For others, it's one of the scariest thoughts to have because it seems impossible to imagine a world where we simply don't exist anymore.

But we aren't here for an existential crisis, we're here to discuss a calculator that can allegedly predict when you're going to meet your end.

size-full wp-image-1263241330
AI can now predict when you are going to meet your end. Credit: Jordan Lye/Getty

Scientists from Denmark and the US have unveiled an artificial intelligence (AI) chatbot, named Life2vec, capable of predicting an individual's lifespan with an astonishing 78% accuracy, as reported by the Metro.

Unlike conventional chatbots such as ChatGPT, Life2vec incorporates data from Danish health and demographic records for six million people, using factors like income, profession, and medical history to make its predictions.

Similar to ChatGPT, the AI model learns to anticipate future events based on the patterns found in the data it processes. However, instead of crafting poetry, Life2vec employs information such as income, profession, and medical records to estimate an individual's life expectancy.

The team utilized Danish health and demographic data, coding various life factors, including gender, smoking habits, mental health, income, and occupational details. Each factor was assigned a specific code, such as S52 for a broken forearm or IND4726 for working in a tobacco shop, which the team then converted into words.

size-full wp-image-1263241332
AI is becoming a major part of the modern world. Credit: Andriy Onufriyenko/Getty

By translating this coded information into data-rich sentences detailing aspects like profession, income, injuries, and pregnancy history, Life2vec generated comprehensive stories about the individuals.

For instance, narratives like: "In September 2012, Francisco received twenty thousand Danish kroner as a guard at a castle in Elsinore" and: "During her third year at secondary boarding school, Hermione followed five elective classes" emerged from this process.

The AI, a large language model (LLM) akin to ChatGPT and Google’s Bard, meticulously pieced together these narratives, functioning like a jigsaw puzzle to predict the expected lifespan of each person.

The team rigorously tested Life2vec on a cohort aged between 35 and 65, achieving an impressive 78% accuracy in predicting who would live and who would die, surpassing traditional models and even excelling in personality test predictions.

If you were hoping to try out the calculator to see when you might meet your end, it's unfortunately (or perhaps fortunately) not available for public use at the moment.

size-full wp-image-1263241333
The death calculator is reportedly 78% accurate. Credit: MR.Cole_Photographer/Getty

Led by Professor Sune Lehmann Jørgensen at the Technical University of Denmark, the research team emphasized that the model's accuracy might be limited to the Danish population, and cautioned against its adoption by insurance companies.

"Clearly, our model should not be used by an insurance company, because the whole idea of insurance is that, by sharing the lack of knowledge of who is going to be the unlucky person struck by some incident, or death, or losing your backpack, we can kind of share this burden," said Professor Jørgensen.

However, he acknowledged the prevalence of similar technologies, suggesting that big tech companies may already be employing comparable models with extensive user data for predictions.

Despite ethical concerns, the team highlighted a potential benefit to the predictive technology - helping researchers discover mechanisms influencing life outcomes and enabling personalized interventions.

Featured image credit: BlackJack3D/Getty

'Scarily accurate' AI calculator can predict when you're going to die

vt-author-image

By James Kay

Article saved!Article saved!

A new AI calculator can apparently predict when you're going to die, and some have described it as "scarily accurate".

Death is something that some people aren't really bothered by and they've already accepted that it'll happen and just get on with their life in the meantime.

For others, it's one of the scariest thoughts to have because it seems impossible to imagine a world where we simply don't exist anymore.

But we aren't here for an existential crisis, we're here to discuss a calculator that can allegedly predict when you're going to meet your end.

size-full wp-image-1263241330
AI can now predict when you are going to meet your end. Credit: Jordan Lye/Getty

Scientists from Denmark and the US have unveiled an artificial intelligence (AI) chatbot, named Life2vec, capable of predicting an individual's lifespan with an astonishing 78% accuracy, as reported by the Metro.

Unlike conventional chatbots such as ChatGPT, Life2vec incorporates data from Danish health and demographic records for six million people, using factors like income, profession, and medical history to make its predictions.

Similar to ChatGPT, the AI model learns to anticipate future events based on the patterns found in the data it processes. However, instead of crafting poetry, Life2vec employs information such as income, profession, and medical records to estimate an individual's life expectancy.

The team utilized Danish health and demographic data, coding various life factors, including gender, smoking habits, mental health, income, and occupational details. Each factor was assigned a specific code, such as S52 for a broken forearm or IND4726 for working in a tobacco shop, which the team then converted into words.

size-full wp-image-1263241332
AI is becoming a major part of the modern world. Credit: Andriy Onufriyenko/Getty

By translating this coded information into data-rich sentences detailing aspects like profession, income, injuries, and pregnancy history, Life2vec generated comprehensive stories about the individuals.

For instance, narratives like: "In September 2012, Francisco received twenty thousand Danish kroner as a guard at a castle in Elsinore" and: "During her third year at secondary boarding school, Hermione followed five elective classes" emerged from this process.

The AI, a large language model (LLM) akin to ChatGPT and Google’s Bard, meticulously pieced together these narratives, functioning like a jigsaw puzzle to predict the expected lifespan of each person.

The team rigorously tested Life2vec on a cohort aged between 35 and 65, achieving an impressive 78% accuracy in predicting who would live and who would die, surpassing traditional models and even excelling in personality test predictions.

If you were hoping to try out the calculator to see when you might meet your end, it's unfortunately (or perhaps fortunately) not available for public use at the moment.

size-full wp-image-1263241333
The death calculator is reportedly 78% accurate. Credit: MR.Cole_Photographer/Getty

Led by Professor Sune Lehmann Jørgensen at the Technical University of Denmark, the research team emphasized that the model's accuracy might be limited to the Danish population, and cautioned against its adoption by insurance companies.

"Clearly, our model should not be used by an insurance company, because the whole idea of insurance is that, by sharing the lack of knowledge of who is going to be the unlucky person struck by some incident, or death, or losing your backpack, we can kind of share this burden," said Professor Jørgensen.

However, he acknowledged the prevalence of similar technologies, suggesting that big tech companies may already be employing comparable models with extensive user data for predictions.

Despite ethical concerns, the team highlighted a potential benefit to the predictive technology - helping researchers discover mechanisms influencing life outcomes and enabling personalized interventions.

Featured image credit: BlackJack3D/Getty