2023-12-19 10:34:52
Updated Tuesday, December 19, 2023 – 09:34
Researchers from the Technical University of Denmark have developed life2vec, an artificial intelligence capable of predicting your death in the next 4 years
SHUTTERSTOCK
An artificial intelligence can analyze data on residence, education, income, health and working conditions to predict events in people’s lives, including estimating the probability of premature mortality or personality nuances.
A study published today by Nature Computational Science shows that if a large amount of data is used to train “transformative models”, they can organize the information and “predict what will happen in a person’s life and even estimate the moment of their death.” .
The project brings together scientists from the Technical University of Denmark and Northeastern University (USA) who analyzed health and labor market linkage data of six million Danes in a model called life2vec.
The team notes that this system is “surrounded by ethical issues” and challenges that must be understood “more deeply before the model can be used,” for example, to assess an individual’s risk of contracting a disease.
After training the model by learning the patterns of the data, they showed that it outperforms other advanced neural networks and predicts outcomes such as personality and the possibility of death “with great precision,” the University of Denmark said in a statement.
The system was able to predict the probability that people in a cohort aged 35 to 65 would survive for four years after January 1, 2016, or capture nuances of personality better than the most advanced models, outperforming them by at least 11%, explained Nature Computational Science.
The paper’s first author, Sune Lehmann, said they had used the model to find out to what extent future events can be predicted based on conditions and events from people’s past.
life2vec predictions are answers to general questions like ‘death in four years?’
The results of the responses are consistent with existing findings in the social sciences; For example, other things being equal, people in leadership positions or with high incomes are more likely to survive.
Life2vec encodes the data into a mathematical structure and the model decides where to place it regarding time of birth, schooling, education, salary, housing and health.
“The exciting thing is to consider human life as a long sequence of eventsin a similar way to how a phrase in a language consists of a series of words,” said the researcher.
Ethical issues
The researchers indicate that life2vec is surrounded “by ethical issues”, such as the protection of confidential data, privacy and the role of bias in data. Furthermore, “it opens up important positive and negative perspectives to be debated and addressed politically.”
These challenges must be understood more deeply before the model can be used, for example, to assess an individual’s risk of contracting a disease or other preventable life events.
Similar technologies for predicting life events and human behavior are already used today by technology companies that track our behavior on social networks, create extremely precise profiles of us, and use them to predict our behavior and influence us.
“This debate must form part of the democratic conversation so that Let’s consider where technology is taking us and if it is an evolution that we want,” said Lehmann.
The next step, according to the team, would be to incorporate other types of information, such as text and images, or about our social connections. This use of data opens – according to the note – a whole new interaction between social and health sciences.
#predict #events #peoples #lives #premature #death