Data transparency & methodology
IPSee does not use random values to create false precision. Every conclusion retains sources, timestamps, confidence and algorithm version; insufficient evidence is reported explicitly.
What the risk score means
Risk ranges from 1–100; higher means more discovered risk evidence. The model separates per-IP evidence from range and ASN reputation baselines. Network baselines are discounted and labelled, and are not individual reports for the current IP. If evidence is insufficient the score is null, not low.
- 1–20 · Low
- 21–50 · Medium
- 51–100 · High
- null · Insufficient data
What confidence means
Confidence combines valid source count, agreement, freshness, direct evidence and manual verification. Conflicts reduce confidence and are disclosed in the professional report.
- 80–100 · High
- 60–79 · Medium
- 40–59 · Low
- 0–39 · Insufficient
Risk model dimensions
Duplicate reports within a dimension use diminishing combination to prevent double-counting. Historical events decay using configurable half-lives.
Abuse, fraud & blocklists30.0%
Proxy, VPN & Tor20.0%
Scanning, spam & attacks15.0%
Bots & automation10.0%
Shared & public exits10.0%
ASN & range stability10.0%
Registry & routing conflicts5.0%
IP classification
Owner type, actual usage and native status are classified independently. ISP ownership does not imply residential use, and residential use does not imply native status.
ownerType · usageType · nativeStatusData handling & privacy
Provider secrets are encrypted and never returned to clients. Browser environment and fingerprint results are not retained by default. Correction contacts are irreversibly hashed and public change records are redacted.
Provider status — last 24 hours
No recent observations yet.
Important limitations
IP intelligence changes with BGP routing, range transfers, proxy exits and provider updates. IPSee results support network assessment and do not guarantee account, payment, ad-review or business outcomes.
Read commercial API docs →