Amlytic
Amlytic is a secure, large scale multi-party transaction data pooling tool for financial institutions. Detecting money laundering patterns in support of fighting financial crime.


Winning tech
Part of DANIE
A community of financial institutions harnessing the latest Privacy-Enhancing Technology (PET) to collaborate and share insights securely, without having to rely on a trusted third party.
Each party keeps control of their own data
No ties to other participants
Easy to use, no IT skills required
Scales to support billions of transactions
Providing large-scale transaction analytics
Collaboration between multiple financial institutions is essential to provide efficient anti-money laundering (AML) and counter-terrorist financing (CFT). But sensitivity of data and regulation prevents actors from pooling data together in clear text.
Scalable to billions of transactions, Amlytic is built on Secretarium's privacy-preserving technology protecting sensitive data, and leverages FutureFlow's large-scale graph analysis to automatically spot irregular relationship patterns and correlations indicative of malicious activity.

Identifies patterns of suspicious behaviour
Pooling demonstrated up to 8x improvement in key metrics
Unsupervised machine learning
Global architecture
Secretarium provides the privacy layers of Amlytic through an encrypted Secure Multi-Party Pseudonymisation (SMPP) engine that securely matches and de-duplicates financial institutions contributions using fuzzy logic. The SMPP engine then uses a hardware secret to generate a fully de-identified pooled dataset.
FutureFlow leverages grid computing to run unique analytics on the de-identified data. The generated networks of transactions are then analysed with artificial intelligence.
Financial institutions are then notified with their original data from the SMPP engine and can leverage Secretarium's Secure Rooms to collaborate, ultimately being able to invite Financial Investigation Units.
1
Data is securely contributed to Secretarium for matching and de-identification
2
De-identified and pooled data is then contributed to FutureFlow for analytics
3
Analytical output is available for proactive and on-demand analysis by financial institutions