PII Detection
Border’s detection engine scans request bodies for 20+ types of personally identifiable information using pattern matching with algorithmic validation.Detection Process
- Quick Check: Fast heuristics determine if content might contain PII
- Pattern Matching: Regex patterns identify potential matches
- Validation: Algorithmic checks confirm validity (Luhn, checksums)
- Confidence Scoring: Each match is assigned high/medium/low confidence
Using the Detection API
Validation Algorithms
Border uses algorithmic validation to reduce false positives:| Type | Algorithm |
|---|---|
| Credit Cards | Luhn checksum + network detection |
| SSNs | Area/group/serial validation |
| IBANs | MOD-97 checksum |
| ABA Routing | Weighted checksum |
| VINs | Check digit validation |
Confidence Levels
| Level | Meaning | Example |
|---|---|---|
high | Algorithmically validated | Luhn-valid credit card |
medium | Strong pattern match | Well-formatted email |
low | Pattern match only | Unformatted 9-digit number |