IP risk is the fraud or compliance risk attached to the internet address from which a customer connects, and it is assessed mainly through IP reputation, the record of how an address has behaved in the past. In payments and banking, IP here means internet protocol, not intellectual property. An IP address seen in earlier fraud, spam or automated attacks is treated as higher risk than one with a clean history.
IP risk is the wider of the two ideas. Reputation is one input; others are the type of connection and its location. Fraud systems ask whether the address belongs to a home broadband line, a mobile carrier, a data centre, a virtual private network, a proxy or the Tor network, because anonymising services hide where a user really is. A geolocation check compares the country of the address with the billing address, the card's country of issue and the customer's usual pattern, and flags connections from sanctioned or restricted territories.
The result is normally expressed as a number that feeds into the overall risk score for a transaction, alongside device fingerprinting and behavioural signals. It is rarely decisive alone, since many customers share addresses and legitimate users also rely on privacy tools.