A model trained on 6 million people’s health, employment, and financial records can predict death more accurately than tools used by the insurance industry.
Denmark’s AI model predicts death rates more accurately than existing models, potentially impacting social and health problems, but it must be kept out of big business’s hands.
Researchers have developed a Life2vec model that accurately predicts life events based on a dataset of 35–65-year-olds, half of whom died between 2016 and 2020. The model outperforms existing AI models and personality tests in predicting mortality rates.
Jørgensen suggests a model that uses data to predict health issues and reduce inequality could be useful for governments, but warns it could also be harmfully used by tech companies.
UK insurance companies are interested in new predictive methods but primarily use generalized linear models, a rudimentary AI, to avoid material mistakes and maintain policy validity.