Three's the Charm: Cloudflare DLP Gets a New PII Record Profile!
Hey everyone, it's me, Shiichan! Today I found a nice little data-protection update to share with you.
Cloudflare ChangelogWhat was announced?
Cloudflare's Changelog announced a new predefined profile for Data Loss Prevention (DLP), called the "PII Record" profile. Just like the name says, it's built to catch records where multiple pieces of personally identifiable information (PII) show up together.
The story so far
Regular DLP detection could fire on a single match, like one email address on its own. That's handy, but it also meant lots of false positives from isolated values that weren't really a leak.
What changes
This new profile only matches when at least three unique detection entries are found close together. So it reacts to spots that really look like a personal record, where things like a name, an address, and a card number line up, instead of buzzing on scattered one-off values. If you want to watch for PII leaks with fewer false alarms, that's a welcome change.
Dive Deep
The profile watches for 14 detection entries:
- AU Passport Number
- American Express Card Number
- Diners Club Card Number
- US Driver's License Number
- Email Address
- Full Name
- US Mailing Address
- Mastercard Card Number
- US Individual Tax Identification Number (ITIN)
- US Passport Number
- US Phone Number
- Union Pay Card Number
- United States SSN Numeric Detection
- Visa Card Number
It only matches once three or more unique entries from this list appear near each other. You turn it on just like any other predefined DLP profile: pick the ready-made one and enable it, no need to build rules from scratch.
Wrap-up
- Cloudflare DLP added a predefined "PII Record" detection profile
- It only matches when three or more unique detection entries sit close together, cutting false positives
- It covers 14 entries, including card numbers, passport numbers, SSNs, and full names
- Because it's predefined, you just select it and switch it on
This one's for the security folks who want to watch for leaks of clustered personal records!