Nigerian lending software company, Lendsqr, is developing an artificial intelligence model designed to assess borrowers’ voices and facial expressions to determine loan eligibility. With a current accuracy rate of 76%, the tool aims to help Nigerians without formal financial documentation access small loans between ₦30,000 and ₦50,000. This innovation could provide a vital alternative to the traditional five Cs of creditworthiness—character, capacity, capital, collateral, and conditions—by enabling lenders to evaluate borrowers through speech and video responses.
Instead of filling out lengthy forms, applicants interact with the AI model by answering questions about their employment and repayment plans via voice or video. The model then predicts the likelihood of loan repayment. Lendsqr is currently funding the pilot project internally and plans to release its findings publicly by Q3 2025, offering the dataset to competitors to bolster lending across the industry. The initiative is particularly focused on Africa, where millions of people remain underbanked due to a lack of documentation.
If successful, the model could significantly expand credit access in Nigeria, where only 6% of adults currently have formal credit, and fewer than 12% of the country’s 41 million small businesses can access loans. By reducing reliance on expensive manual verification, the AI tool could lower lending costs and increase profitability for lenders. Lendsqr’s clients, including Snapcash, Blockacash, and Kredi, are expected to benefit from broader market reach and better screening.
Backed by funding from the Nigerian Ministry of Communications, Innovation & Digital Economy and Google, the AI model is slated for full rollout once it achieves 90% accuracy. While it may not replace traditional lending for high-value loans like mortgages or car financing, it promises to democratize access to foundational credit—offering small but potentially life-changing loans to underserved Nigerians and potentially to immigrants in markets like Canada.