Labol completes AI fraud detection model with University of Tokyo

Labol Inc. has completed the design and development of a new fraud detection model for its AI-driven invoice factoring service, ‘labol’. The project was conducted in collaboration with Professor Kenji Tanaka of the University of Tokyo’s Graduate School of Engineering.

The research, which began in March 2024, focused on developing a machine learning model to control uncollected payment risks. The new system is designed to automatically detect suspicious data by training AI on characteristics of previously forged invoices, replacing manual visual inspections.

By integrating this model into the ‘labol’ screening system, the company aims to prevent fraudulent activities while simultaneously increasing the speed of the screening process. This development is intended to provide faster financial support for freelancers, sole proprietors, and small-to-medium businesses (SMBs). The project also involved support from Delight Flow Inc., a startup led by a member of Professor Tanaka’s laboratory, and Albacross Inc., a venture capital firm and shareholder of Labol.