evidoria

← Back to browse

Good practice Imported

JAMB's Biometric and AI Facial-Recognition Verification for Nigeria's National University Entrance Exam

Nigeria · Abuja · See the Nigeria profile · See the Abuja profile

Evidence: Observational / pre–post Top 95% 20/100 · Ask Evidence Copilot about this practice

Nigeria's JAMB uses AI facial recognition plus fingerprint biometrics to verify over 2.2 million UTME candidates a year and block exam impersonation, but verification failures affected ~80,000 candidates in 2023 and fraudsters now use AI face-blending to defeat it.

2,243,816 candidates
UTME candidates registered (2026 cycle) (2026)
~80,000 candidates
Candidates affected by biometric verification failures requiring mop-up exam (2023)
1,787 candidates
Candidates falsely declaring albino status (2025)
JAMB's Biometric and AI Facial-Recognition Verification for Nigeria's National University Entrance Exam

Details

Maturity
Established
Promoter
Joint Admissions and Matriculation Board (JAMB)
Period
2015–ongoing (AI facial-recognition checks intensified 2023–2026)
Keywords
exam integrity, biometric verification, facial recognition, higher-education admissions, fraud prevention

Context

JAMB administers Nigeria's UTME and deploys fingerprint and AI-based facial-recognition verification at Computer-Based Test centres nationwide to block candidate impersonation ('mercenary' exam-taking). JAMB registered 2,243,816 candidates for the 2026 UTME.

Results

In 2023, roughly 80,000 candidates were affected by biometric verification failures (mismatches, non-recognition, other errors), forcing a rescheduled 'mop-up' exam; similar failures recurred in the 2025 mop-up exams. JAMB's registrar, Prof. Is-haq Oloyede, described fraudsters using AI to blend two people's faces to defeat facial recognition. In 2025, 1,787 candidates falsely declared themselves albino — suspicious given genuine albino Nigerians would number under 250 a year — apparently to exploit looser facial-verification thresholds.

Conclusions

This is best read as a mixed-evidence, cautionary case: the system operates at massive national scale but has not eliminated fraud and has repeatedly failed to verify genuine candidates, illustrating the operational fragility of AI biometrics deployed nationally without commensurate error-handling capacity.

Implementation
Indicative cost
High (€500k–€5M) — National biometric verification infrastructure covering over 2.2 million candidates per cycle; no specific cost figures disclosed.
Time to results
Long (> 3 years) — System in place since ~2015, with AI facial-recognition checks intensified 2023–2026.
Staffing & skills
JAMB registrar Prof. Is-haq Oloyede, JAMB Head of Public Affairs and Protocol Dr. Fabian Benjamin

Conditions for success

  • Fingerprint plus AI facial-recognition verification deployed at CBT centres nationwide

Common failure modes

  • Recurrent biometric verification failures requiring rescheduled 'mop-up' exams (~80,000 candidates in 2023, recurrence in 2025)
  • Fraudsters using AI to blend faces to defeat facial recognition
  • Suspicious over-declaration of albino status (1,787 in 2025) apparently to exploit looser verification thresholds
  • Operational fragility of AI biometrics deployed nationally without commensurate error-handling capacity

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful