Every transaction processed through Privex Pay is evaluated by an AI-powered fraud detection system before it is authorized. The system operates in real time, analyzing hundreds of behavioral and contextual signals simultaneously — with no impact on checkout speed for legitimate customers. This page explains how the detection works, what triggers a flag or block, and what your obligations are as a merchant.Documentation Index
Fetch the complete documentation index at: https://privexpay.mintlify.app/llms.txt
Use this file to discover all available pages before exploring further.
How fraud detection works
Privex Pay’s fraud engine evaluates each transaction the moment it is submitted. The system does not rely on a single rule or threshold. Instead, it builds a risk profile for each transaction by combining multiple layers of analysis. What the system analyzes:- Transaction patterns — velocity checks, unusual purchase amounts, frequency of transactions from a single card or account, and deviations from a customer’s typical behavior.
- Behavioral signals — how a user interacts with the checkout flow, including timing, navigation patterns, and input behavior that can distinguish humans from bots.
- Device and network context — device fingerprinting, IP reputation, use of VPNs or proxies, and geographic inconsistencies between the billing address and the network location.
Real-time alerts
When a transaction is flagged, Privex Pay sends a real-time alert to your account. You can review flagged transactions in your dashboard and, where applicable, take action. Blocked transactions are not charged to the customer.Fraud alerts appear in your Privex Pay dashboard under Transactions. You can configure notification preferences in your account settings.
Customizable fraud rules
In addition to the default AI-powered detection, you can configure supplementary fraud rules for your account. These rules let you apply additional controls based on factors specific to your business — such as blocking transactions from certain countries, flagging orders above a defined amount, or limiting the number of transactions per card within a time window.What happens when fraud is detected
Transaction is evaluated
The fraud engine scores the transaction in real time as the customer submits payment.
High-risk transactions are blocked or flagged
Transactions that exceed the risk threshold are either blocked immediately (no charge is made) or flagged for manual review, depending on the risk score and your configured rules.
You receive an alert
Privex Pay sends a real-time notification to your account. Review the flagged transaction in your dashboard.
Your obligations as a merchant
Privex Pay’s fraud detection protects your business, but it depends on merchants operating honestly and in good faith. You must not:- Submit fraudulent, synthetic, or fictitious transactions
- Use bots, scripts, or automated tools to generate artificial transaction volume
- Encourage or facilitate friendly fraud (customers disputing legitimate charges)
- Misrepresent products or services in a way that predictably leads to disputes
Excessive chargebacks
Chargebacks occur when a customer disputes a charge with their card issuer. While some chargebacks result from genuine fraud, many stem from unclear product descriptions, unmet expectations, or failure to issue timely refunds. If your chargeback rate exceeds acceptable thresholds, Privex Pay will contact you and may take the following steps:- Require a reserve fund to cover potential future chargebacks
- Impose transaction velocity limits
- Suspend payout processing pending review
- Terminate your account if the rate is not resolved