Cracking the Credit Code: Alternative Data and AI for Financial Inclusion
Traditional credit scoring systems often exclude women and underserved borrowers – especially those without formal financial histories or collateral – and this limits their access to credit and ability to grow their businesses and improve their livelihoods. In response, a new generation of credit scoring models powered by alternative data and artificial intelligence (AI) is redefining credit scoring for borrowers who lack formal credit histories. Yet, the benefits of this shift are uneven – women, informal workers, and microentrepreneurs remain underrepresented in credit portfolios.
This report aims to inform practitioners and policymakers about ways in which innovations in credit scoring can advance financial inclusion while ensuring fairness and protecting consumers. It was guided by three central issues:
- Market Trends: How are alternative data and artificial intelligence used in credit scoring, particularly across emerging markets?
- Gender Inclusion: To what extent do these models expand women’s access to formal credit?
- Data and Model Design: What types of data and modeling approaches underpin these systems, and how might they differ for women borrowers?
The analysis draws on interviews with more than 30 experts in fintech and credit scoring and a global mapping of 448 alternative-credit scoring firms. It also includes a literature review and borrower-and customer-level insights from the fintech companies Eshandi in Zambia and Vexi in Mexico. Combined, these capture diverse data use, business models, and regional operations, while highlighting emerging partnerships, regulatory developments, and challenges around fairness and transparency.
