Artificial Intelligence and Algorithmic Gender Bias in African Fintech Lending Platforms

African Journal of Gender and Women Studies | Vol. 11, No. 9, August 2026 | pp. 57–64

DOI: 10.46882/2026/AJGWS/001255

Article Type: Original Research Article

Title: Artificial Intelligence and Algorithmic Gender Bias in African Fintech Lending Platforms

Names of Authors: Dr. Chinedu U. Okafor¹, Prof. Fatoumata B. Diallo²

Authors’ Affiliations: ¹Department of Computer Science and Gender Studies, University of Nigeria, Nsukka, Nigeria; ²African Center for Technology Studies, Dakar, Senegal

Abstract:
This study investigates the presence and impact of algorithmic gender bias within artificial intelligence (AI) driven credit scoring models used by fintech lending platforms across West Africa. Utilizing a mixed-methods audit methodology, we evaluated the performance of three major commercial credit-scoring algorithms using an anonymized dataset of 45,000 loan applications. The quantitative analysis reveals that female applicants are assigned a 28% lower creditworthiness score on average compared to male counterparts with identical income-to-debt ratios (p < 0.01). This disparity stems from predictive variables that penalize fragmented employment histories and informal transaction patterns, which are highly prevalent among female entrepreneurs. Furthermore, the machine learning models utilize historical proxy variables that inadvertently perpetuate systemic patriarchal exclusions in formal asset ownership. Qualitative focus groups with female micro-entrepreneurs revealed that automated rejections lack transparent explanations, leaving applicants without recourse or understanding of how to improve their credit profiles. This algorithmic exclusion limits women's access to digital capital, widening the digital gender divide in the region's expanding digital economy. The study concludes that algorithmic neutrality is a myth when trained on historically biased data. We propose an alternative fairness-aware machine learning framework that integrates gender-disaggregated financial indicators to guarantee equitable credit distribution in African fintech ecosystems.

Keywords: Artificial Intelligence, Algorithmic Bias, Fintech Loans, Financial Inclusion, Machine Learning, West Africa

Manuscript Timeline: Received: March 12, 2026; Revised: May 20, 2026; Accepted: June 18, 2026; Published: August 05, 2026.

Citation: Okafor, C. U., & Diallo, F. B. (2026). Artificial Intelligence and Algorithmic Gender Bias in African Fintech Lending Platforms. African Journal of Gender and Women Studies, 11(9), 57–64. DOI: 10.46882/2026/AJGWS/001255

Subscribe to read the full article: https://internationalscholarsjournals.org/subscribe-to-read