As surprising capabilities of artificial intelligence (AI) emerged this year, one of the questions that people ask is whether AI could be used to predict life events.
Now, a new study shows that AI models can analyse data on people's residence, education, income, health and working conditions and predict life events with high accuracy.
The study, led by the Technical University of Denmark (DTU) and three other universities, shows if a large amount of data about people's lives is used to train AI models such as OpenAI’s ChatGPT, they can systematically organise the data and predict what will happen in a person's life – even estimate mortality. The findings were published in the journal Nature Computational Science.
"We used the model to address the fundamental question: to what extent can we predict events in your future based on conditions and events in your past? Scientifically, what is exciting for us is not so much the prediction itself, but the aspects of data that enable the model to provide such precise answers," first author Sune Lehmann said in DTU’s press statement.
When the new model, Life2vec, was asked questions about mortality, the answers were consistent with the findings within social sciences, the researchers explained. For instance, if all things are equal, people in a leadership position or with a high income are more likely to survive.
The model considers human life as a long sequence of events, similar to how a sentence in a language consists of a series of words, Lehmann explained in the press statement. After initial training, wherein the model learned the patterns in the data, it outperformed other advanced neural networks and predicted outcomes such as personality and mortality with high accuracy.
However, it’s important to note that there are both positives and negatives to such advancements. Currently, similar technologies for predicting life events are being by tech companies to profile people accurately, and use these profiles to predict our behaviour, Lehmann points out in the statement. Hence, conversations about the use of such technology should be more democratic so that there is transparency in deciding if people want this kind of development in the first place.