top of page

Privacy Enhancing technology applications

Our Services

Telecom Fraud detection

In telecom fraud detection, federated learning can be employed to develop robust models for detecting fraudulent activities across various telecom networks. By allowing individual telecom operators to train models locally on their own data without sharing sensitive information, federated learning enables the creation of a global fraud detection model that benefits from the diverse data across different networks while preserving privacy.

Banking fraud detction

In banking, federated learning can be applied to build fraud detection models collaboratively. Different banks can contribute insights from their customer transactions without sharing personal data. This helps in creating a more generalized and effective fraud detection model that can adapt to evolving fraud patterns while respecting privacy regulations.

Threat Intelligence sharing

Using Secure Multi-Party Computation or SMPC can enable secure computations on shared threat intelligence data. Organizations can jointly analyze and identify emerging threats without compromising the confidentiality of their individual threat data.

IoT security

In IoT security, federated learning can be used to improve anomaly detection and security models across a network of connected devices. Each device can locally train its model based on its specific environment, contributing to a global model that enhances the overall security posture without exposing individual device data.

BT Group collaborates with PryvX

Demoing a privacy enhancing platform at BT’s innovation showcase in Adastral Park, PryvX and BT are working together to help to combat cyber-crime and fraud in financial and telecom sectors

What is Federated Learning?

Federated Learning is a decentralized machine learning approach that enables model training across multiple devices or servers holding local data, without exchanging them centrally.

bottom of page