Fraud
Industry News
The trials involved ten million artificial transactions and demonstrated that collaborative models were twice as effective in real-time fraud detection compared to models trained on data from a single institution.
Key highlights from the pilot include:
➡️ Secure data sharing using PETs while preserving privacy and security;
➡️ Real-time verification of suspicious accounts and detection of anomalous activity;
➡️ Use of federated learning to train AI models locally without sharing customer data; and
➡️ A scalable framework for industry-wide fraud defence.
✅ Financial institutions should explore collaborative approaches to fraud detection, leveraging emerging technologies to reduce risk and operational costs.
💡 Plenitude’s AI Assurance Service supports clients in navigating AI risk through robust control assessments and governance frameworks.
Learn more: https://www.plenitudeconsulting.com/services/ai-assurance-services
