University of Tokyo Researchers Develop AI Tool that Forecasts Employee Turnover

Bosses who are concerned about employee turnover or are unsure of how long their employees will stay with the company will soon have access to artificial intelligence (AI) technology that can predict which employees are likely to leave next. Developed by Japanese researchers, this tool is designed to help companies provide targeted support to their employees in order to prevent them from quitting.

Employee data, ranging from work attendance to personal details like age and gender, is analyzed using this tool created by a professor from Tokyo City University in collaboration with a local startup. In addition to current employees, the tool also looks at data from former employees who have left the company in the past.

Professor Naruhiko Shiratori explains that from this data, the system can predict the resignation rate of new hires as a percentage. The AI tool is currently being tested in several companies to create customized models for each of them.

The results provided by the tool can be used by managers to have conversations with high-risk employees about potential support the company can offer. Rather than showing employees raw data, which may be surprising or unwelcome, the AI predictions give managers a way to address potential difficulties the employee may be facing.

This AI tool builds on previous research that used AI to predict which college students were likely to drop out. In Japan, companies typically hire new graduates each April, but a significant percentage leave their jobs within the first year, and even more after three years, according to government data. Japanese companies are increasingly focusing on supporting their young employees, especially given the country’s aging population and resulting labor shortages in key industries.