AI in the modern landscape is multifaceted, and its development is driven by various factors. Many fields utilize AI technology to automate repetitive tasks, and processes, and analyze massive data sets. AI can also be used to generate new and innovative ideas in the creative scene.
Current AI tech has already proven to be an effective tool that can assist human operations in our daily lives. Still, new developments in technology show that it may be capable of doing more than what is expected.
For some, predicting the future is something that borderlines pseudoscience, but researchers from Denmark have developed AI technology that can ‘foresee’ life events and life spans based on a clear database. The AI tool, called Life2vec, is said to be capable of predicting just about anything vital included in a person’s life based on their education, financial capacity, health history, and job experiences. The results generated by the groundbreaking AI tool are all based on the data collected from the entire population of Denmark.
Researchers Developed a New AI That Can Predict Life Events
Image from Northeastern Global News
From an individual’s personality to their possible lifespan, the AI tool Life2vec is said to be capable of providing highly accurate predictions about such factors. According to the study’s lead author, Sune Lehmann, the predictions generated by Life2vec are significant on their own, since the future-predicting AI tool utilizes true life events and patterns procured from actual people living in Demark.
"This model offers a much more comprehensive reflection of the world as it is lived by human beings than many other models," said Lehmann. "The whole story of a human life, in a way, can also be thought of as a giant long sentence of the many things that can happen to a person.
Lehmann, a professor at the Technical University of Denmark, highlights the process of Life2vec and how it was developed to become fully equipped to create accurate human life predictions. Data collected from the population of Denmark was used by the AI tool to study millions and millions of life event sequences. Through these sequences, vector representations are created in 'embedding spaces,' serving as the foundation for Life2vec to organize and establish categorized connections between various life events and factors such as education, income, and health.
As a result, Life2vec has trained itself to generate accurate life sequences based on actual, real-life situations. Predicting human mortality is now no longer just a possibility, it was shown to be possible through Life2vec.
"When we visualize the space that the model uses to make predictions, it looks like a long cylinder that takes you from low probability of death to high probability of death," Lehmann said. "Then we can show that in the end where there's a high probability of death, a lot of those people actually died, and in the end where there's [a] low probability of dying, the causes of death are something that we couldn't predict, like car accidents."
According to Northeastern University professor Tina Eliassi-Rad, the use of this AI tool still has limitations, both on its capacity and where it should be utilized, considering that Life2vec’s data only relies on the residents of Denmark and does not include a broader spectrum of life sequences.
“Even though we're using prediction to evaluate how good these models are, the tool shouldn't be used for prediction on real people,” said Eliassi-Rad. "It is a prediction model based on a specific data set of a specific population."