Blog

Residential proxies are now the default fraud infrastructure

Industry data in 2026 shows residential proxies have become the majority of fraud traffic’s infrastructure. Why static IP blocklists are finished, and what replaces them.

One of the clearest signals in the 2026 fraud data is also one of the least discussed outside the industry: residential proxies have gone from a niche tool to the default infrastructure behind online fraud. By some measurements, residential and mobile proxy traffic now makes up the large majority of fraud attempts, up sharply year on year, and analysts are calling the shift structural rather than seasonal. For anyone still relying on IP blocklists, this is the quiet end of an era.

What a residential proxy actually is

A residential proxy routes an attacker's traffic through a real consumer's internet connection — a home broadband line, a phone, sometimes a device whose owner has no idea it is being used. To your defences, the request arrives from an ordinary residential address in a plausible city, carrying none of the obvious marks of a datacenter. That is the entire value proposition the proxy sellers advertise: traffic that looks exactly like a real customer.

Why static blocklists lose

The old model — maintain a list of bad IPs and block them — assumed bad traffic came from a stable, identifiable set of addresses. Residential proxy networks destroy that assumption:

  • The pool is enormous and rotating. Millions of residential IPs, cycling constantly, so the address you block today is a real customer's tomorrow.
  • Blocking the IP hits real users. Because these are genuine consumer connections, a blocklist entry risks turning away the actual person whose line was borrowed.
  • Lists are always stale. A blocklist describes where fraud was last week. Rotation means it never catches up.

I have watched teams pour effort into ever-bigger blocklists and lose anyway, for exactly these reasons. (We wrote about the mechanics in why blocklists fail against residential proxies.)

What replaces the list

You cannot win this with a static list; you win it with live classification and context. Instead of asking "is this specific IP on a bad list?", ask "what kind of network is this, right now, and does its behaviour fit a real customer?" A residential proxy still has characteristics — the way the network is operated and classified, the mismatch between a claimed location and the rest of the signals, the velocity of requests across a pool — that a live service can read even when any single IP looks clean. Combine that with corroborating signals like the card's issuing country, and the proxy's disguise stops working.

Classify live, don't list

Fraudex classifies any IP at the moment of the request — network type, proxy and VPN detection, a bot verdict — rather than checking it against a list that went stale the day it was built. See the proxy detection API and the VPN detection API, or test an address yourself with the IP checker.