Transaction context helps detect threats
Device, behavior and client history add to transaction analysis in cases of social engineering, phishing and SIM swap.
Rules, ML scoring and behavioural biometrics to detect suspicious transactions in real time.
Antifraud is now a bank’s obligation
Kazakhstan. In a year and a half, the NBK Antifraud Center has brought together more than 270 participants and stopped transactions worth ₸127 bn, and banks now reimburse transfers to blacklisted accounts, so transaction risk must be assessed before execution, not investigated afterwards. Uzbekistan. Fraud is card-based: in six months banks stopped 18 M suspicious transactions, and the Central Bank is proposing mandatory 24/7 antifraud with quarterly model reviews.
The antifraud platform checks transactions in real time based on the client’s device, behavior and history
Device, behavior and client history add to transaction analysis in cases of social engineering, phishing and SIM swap.
Decision history and reporting help investigate incidents and respond to regulator requests.
False positives get in clients’ way. Thresholds are tested on historical data and refined using analyst labeling.
Scenarios for phone fraud, SIM swap and cash-out. Business rule editor with no releases.
Real-time assessment of anomalies in amount, recipient, device and behavior.
Analysis of typing, swipes and session dynamics to detect signs of account takeover.
Device and link analysis: emulators, VPNs, device farms and account reuse.
Prioritized alert queue with SLAs and auto-filled investigation reports.
Exports and reports for the risk committee, auditors and regulators; decision history.
Events collected from mobile banking, digital banking, processing and support via SDK and API.
Event normalization, linking of device, geolocation, recipient and available client history.
Rules and ML decide the action: allow, step up authentication or stop the transaction.
Actions under the risk policy, incident handover to the security team, retraining on analyst labeling.
“Antifraud is not about blocking, it’s about trust: clients shouldn’t notice the protection until they need it. That’s why we start with the bank’s historical data, not with off-the-shelf rules.”
Oleg SolabayevCTO, Head of Product
First come testing on historical data, threshold tuning and a rollout plan. Connection to the production environment and timelines are set out in the contract.
Yes. Events come from the bank’s systems, and the platform is deployed in the bank’s infrastructure.
The set of additional features depends on the existing fraud monitoring system. The platform connects on top of it: ML scoring, behavioral biometrics and investigations.
Scenarios and reporting are agreed for the bank’s jurisdiction. Decision history and exports are used in incident reviews and responses to the regulator.
Your request goes to our business development team: we will get in touch, work out what needs to go live at your bank and propose a solution.
Thank you, your request has been sent
A business development manager will contact you using the details you provided.
