Catalog record · Fraud Detection & AML
Stage as of Jun 24, 2026 · historical report, not live reconfirmation.
KakaoBank joint federated-learning fraud model
Primary workflow source. A named organization's workflow supported by a reviewed primary source. Stage is reported separately; this label does not establish measured outcomes or independent validation.
Last researched Oct 26, 2026. Source access and reported stage are recorded below; deployment evidence does not establish ROI.
What the research supports
Financial Security Institute reports a tested joint voice-phishing detection model with KakaoBank, Toss Bank and Kbank, sharing model weights across institutions.
- Reported stage
- Pilot
- Technology
- Machine Learning
- Business function
- Middle Office
- Impact (legacy label)
- Not labeled
- Catalog date · unconfirmed
- Not reported
- Imported
- Oct 29, 2026
- Last researched
- Oct 26, 2026
Measures & evidence
No quantified measure recorded. This is a gap in the evidence, not a finding of no value.
Documented vendor relationships
No vendor relationship has been linked to this specific workflow in the curated overlay. This does not mean it was built without vendors.
See the company’s documented relationshipsDeployment evidence & limits
Joint model tested; operational use planned for July 2026. Status as of 2026-06-24.
Stage is as reported in the cited evidence. Current operating status has not been independently reconfirmed; the research date is an editorial review date.
June release does not confirm planned July production occurred.
Recorded dependencies
- Participating banks' existing fraud systems and federated model sharing.
Source register
- 금융보안원-인터넷은행 3사 카카오뱅크, 토스뱅크, 케이뱅크 국내 최초 연합학습 기반 보이스피싱 탐지 AI 공동모델 적용 실시fsec.or.kr · opens in a new tab
Financial Security Institute · issuer · full-text · Published 2026-06-24
Research lanes: fintech_lending · Batch global-discovery-2026-10-26. Coverage and remaining gaps