AI-powered KYC address verification
for Nigerian fintechs
Compliance officers manually match identity documents against utility bills across thousands of onboardings. This tool does the cross-check in seconds and returns a structured verdict with auditable reasoning.
Identity Document
Nigerian NIN slip or driver's license
Drop file here or click to upload
JPG or PNG, max 10 MB
Proof of Address
Utility bill (PHCN, IKEDC, DSTV) or bank statement
Drop file here or click to upload
JPG or PNG, max 10 MB
Or try a demo scenario
How it works
Upload documents
Upload a Nigerian identity document (NIN slip or driver's license) and a proof of address (utility bill or bank statement).
AI extracts data
Claude vision reads both documents and extracts structured fields — name, address, date, and issuer — without storing any data.
Cross-check & verdict
Four deterministic checks run against the extracted data. The AI then writes a compliance verdict with explicit, auditable reasoning.
What gets checked
Name match
Compares the full name on the ID against the name on the proof of address. Handles abbreviated first names, token reordering, and common Nigerian naming patterns.
Address legibility
Verifies the proof-of-address document contains a complete, parseable address with enough components (street, area, state) to be actionable for compliance.
Document recency
Confirms the proof of address was issued within the last 90 days, matching CBN requirements for KYC tier verification.
Authenticity signals
Flags AI-detected anomalies — blurry critical fields, font inconsistencies, unusual document layouts, or signs of digital tampering.
Why this matters
The Central Bank of Nigeria's KYC framework requires fintechs to verify customer addresses before granting Tier 2 and Tier 3 account privileges — higher transaction limits, lending products, and cross-border transfers. At scale, this means compliance teams manually comparing thousands of identity documents against utility bills daily: a process that is slow, expensive, and introduces human error into regulated decisions.
This tool demonstrates how AI vision can automate the extraction and cross-check step, producing a structured verdict with explicit reasoning in under 10 seconds. The deterministic check layer — name match, address legibility, document recency, authenticity signals — ensures every decision is auditable and explainable, a critical requirement for any compliance workflow operating under CBN oversight.