JapanFinance and insurance2023 – ongoingSBI Shinsei Bank
AI scoring for suspicious-transaction investigations
SBI Shinsei Bank reported about 50% fewer transaction investigations after introducing AI scoring
Problem
Suspicious-transaction monitoring required investigators to distinguish transactions needing attention from those requiring no further inquiry. Growing transaction volumes added to that workload.
Approach
The bank introduced an internally developed AI scoring model in 2023 and automated data integration in 2024. The model supplied suspicion scores and explanations. Updated training data helped it reflect changing financial-crime patterns.
Results
Takeaway
The bank reduced investigation volume by filtering transactions judged unnecessary to investigate. It did not disclose a false-negative rate.
Categories
- Size
- Large enterprise
- Industry
- Finance and insurance
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
The bank reported about 50% fewer investigations while substantially maintaining monitoring coverage. It described model retraining and the earlier introduction of automated data integration.
Full details are in the original sources.
