
RESOURCES / CASE STUDIES & SOLUTION BRIEFS
Solution Brief: Online Banking Scams
Overview: Online Banking Scams
- Fraud losses from scams are overtaking fraud losses from account takeovers.
- Scams are hard to detect because they are initiated by the legitimate customer.
- Darwinium combines machine learning with transaction intelligence, behavioral biometrics and victim propensity modelling to alter potential scams journeys in real time
Internet Banking Scam Detection & Prevention
Stop Authorised Push Payment (APP) Fraud in Real Time
Fraud losses from scams are now overtaking account takeover losses. Banks need a new approach. Darwinium helps financial institutions detect, disrupt, and prevent internet banking scams as they happen—without disrupting legitimate customers.
Protect customers from scams. Reduce APP fraud losses. Act in real time.
The Growing Threat of Internet Banking Scams
Authorised Push Payment (APP) scams are one of the fastest-growing fraud threats in digital banking. Unlike traditional fraud, these scams are initiated by legitimate customers, making them significantly harder to detect.
Why scams are increasing:
- Fraudsters are shifting from account takeover to social engineering attacks
- Instant payments enable rapid movement of stolen funds
- Customers are targeted using highly personalised deception tactics
In mature digital banking markets, scam losses are now surpassing traditional fraud categories.
Why APP Fraud Is So Hard to Detect
Traditional fraud tools weren’t designed for scams. Here’s why:
1. Customer-initiated transactions
Customers complete payments themselves, often passing authentication checks like 2FA.
2. Sophisticated social engineering
Fraudsters impersonate banks, governments, and trusted entities using advanced tactics like spoofed calls and fake support teams.
3. Faster payments ecosystem
Funds are transferred and dispersed across mule accounts within minutes, making recovery nearly impossible.
The Darwinium Approach to Scam Prevention
Darwinium goes beyond detection. We intervene in the customer journey in real time to stop scams before money leaves the account.
Core capabilities:
- Real-time scam detection
- Dynamic customer journey orchestration
- Continuous behavioral and transaction monitoring
- Victim propensity modelling
Darwinium combines machine learning, behavioral biometrics, and transaction intelligence to identify scam signals and act instantly.
How It Works: Real-Time Scam Disruption
Instead of relying on static rules or post-transaction alerts, Darwinium actively alters high-risk customer journeys.
Example interventions:
- Real-time warnings tailored to the specific transaction
- Payment friction introduced only when risk is detected
- Prompts encouraging customers to verify with their bank
- Adaptive journeys based on evolving risk signals
These interventions are contextual, personalised, and effective, reducing scam success rates without harming UX.
Key Features for Internet Banking Scam Detection
Dynamic Customer Journey Management
- Real-time, session-specific scam warnings
- Tailored messaging to prevent “alert fatigue”
- No dependency on legacy banking infrastructure
Machine Learning for Scam Detection
- Models trained specifically on scam behaviours, not just fraud
- Detection of unusual login patterns, new beneficiaries, and transaction anomalies
- Continuous learning based on evolving scam tactics
Transaction Intelligence
- Beneficiary risk scoring (including mule account detection)
- Payment anomaly detection based on user behaviour
- Cross-session and cross-channel analysis
Behavioral Biometrics & Victim Modelling
- Detect hesitation, stress, or unusual device usage
- Identify vulnerable or previously targeted customers
- Recognize signs of coercion or social engineering in progress
Built for the Modern Banking Landscape
As instant payments adoption increases globally, scam volumes are rising across:
- UK and Europe
- US real-time payments ecosystem
- Emerging digital banking markets
Regulators are also increasing pressure on banks to reimburse victims and improve protections, making proactive scam prevention essential.
Highlights
- UK
One of the first European countries to adopt faster payments and with strong market penetration of online and mobile banking.
- Mainland Europe
Several European banks are currently between stages two and three of their journeys. One Italian bank saw fraud increase 8,000% in one month after introducing instant payments.
- United States
Some scams are still classified as first-party fraud, but we would expect similar increases in scams volumes as instant payments are rolled out.
Online Banking Scams
Download the Solution Brief