When AI predicts employee resignations

(Tokyo) Bosses worried about turnover in their workforce or wondering how long a recruit will stay in their job could soon count on artificial intelligence (AI) to know the next employee likely to leave their company.


Such a tool was developed by Japanese researchers to help companies provide targeted support to their employees, in order to dissuade them from resigning.

Employee data, from their work attendance to personal information such as age and gender, is analyzed by this tool created by a Tokyo City University professor with start-up local.

The tool also analyzes data on former employees who have left the company.

From all this data, the tool predicts the resignation rate of new recruits as a percentage, explained Professor Naruhiko Shiratori.

“We are in the testing phase with this AI tool in several companies, creating a model for each of them,” added the researcher.

Bosses could use the results to “suggest to the employee at high risk (of resignation, Editor’s note), without showing him the raw results which could shock him – that the company is ready to offer him support, because the AI ​​predicts that “He could face difficulties,” Shiratori said.

To create this tool, the researchers relied on a previous study that used AI to predict the profile of university students likely to drop out.

Japanese companies traditionally hire young graduates at the same time each year, in April.

But about 10% of these recruits leave their jobs in their first year, and about 30% within three years, according to government data.

Japanese companies are increasingly seeking to take care of their young employees, against a backdrop of the accelerated demographic decline of the Japanese archipelago, which is creating labor shortages in many sectors of activity.


source site-55