Nigerian fintech Mathesis Analytics has secured a strategic investment from Sewa Capital to scale its AI-powered credit decisioning engine, bringing a platform that has already supported more than 8 million loans deeper into institutional lending. The investment amount was not disclosed.
Mathesis addresses what its founders identify as a structural gap in Nigerian credit markets: verifiable consumer data is fragmented across systems, making it difficult for lenders to price risk accurately at scale. The company aggregates disparate transactional and behavioural data into a real-time creditworthiness measure that financial institutions can embed in their own underwriting.
The funding will support product development, institutional integrations, expansion of the lender network, and strengthening of data and technology infrastructure.
Operating base
The traction figures are the most substantive part of the announcement. Mathesis says it has supported more than 8 million loans for over 2 million unique borrowers in Nigeria — an operating history that gives the company both a commercial foundation and, more importantly for a credit model, a repayment dataset to train on.
“Credit inclusion begins with information visibility,” said Winston Osuchukwu, founder and chief executive of Mathesis Analytics. Partnering with Sewa Capital, he said, accelerates the rollout of the infrastructure needed to equip a wider network of lenders to evaluate risk accurately and extend credit to previously excluded populations.
Angela Jide-Jones, managing director at Sewa Capital, framed the deal as an infrastructure bet. “We are interested in businesses building the infrastructure that enables African economies to allocate capital more efficiently and inclusively,” she said, describing the information gap as a fundamental constraint on lenders’ ability to assess risk confidently.
The competitive picture
Mathesis enters a segment where the scale gap is considerable. Optasia, the AI-led fintech that listed on the JSE in November, processes more than 32 million near-instant credit decisions daily across 121 million active users, has extended over $20 billion in small loans, and reported a 1.2% default rate for 2025 — with Nigeria and South Africa among its largest markets.
The alternative-data credit scoring category is also getting crowded at the earlier stage. Nigerian startup Regxta and Angola’s Anda Africa both graduated from Google’s tenth Accelerator Africa cohort in June, applying AI credit scoring to unbanked micro-businesses and informal moto-taxi workers respectively.
Mathesis’s positioning as infrastructure sold to lenders, rather than a lender itself, differentiates it commercially — the company grows with adoption across institutions rather than by carrying loan-book risk.
The regulatory question
That business model runs directly into legislation currently before Nigeria’s National Assembly.
The Artificial Intelligence Control and Regulation Bill explicitly identifies credit scoring as a high-impact decision category. Under its provisions, individuals subject to automated decisions in credit must be informed that a system is involved and given a route to challenge the outcome. Developers would need to document system design and data sources for council review, and maintain audit trails ensuring explainability and traceability. The bill also aligns with Nigeria’s Data Protection Act 2023, requiring that personal data used in AI training be processed lawfully and without discrimination on grounds including gender, race or religion.
Those requirements bear specifically on behavioural-data credit models. The value of alternative data is that it finds signal where formal credit history does not exist. The risk is that behavioural proxies can correlate with protected characteristics in ways that are difficult to detect and harder to explain to a borrower who has been declined.
The Central Bank of Nigeria has separately signalled it will require regulated institutions to adopt AI-based compliance and real-time reporting systems within three years, as part of the Nigeria Payments System Vision 2028.
For Mathesis, building explainability and audit capability into the engine now is likely to be cheaper than retrofitting it. It may also become a commercial advantage: lenders integrating third-party decisioning infrastructure will increasingly need to demonstrate to regulators that they can account for how a decision was reached.
