Kenyan banks are increasingly deploying artificial intelligence and alternative data in credit risk management as the sector confronts elevated levels of non-performing loans and seeks more accurate ways to assess borrowers.
According to the Kenya Bankers Association, lenders are integrating AI across the credit lifecycle, particularly during loan appraisal, where models can analyse information from multiple sources to generate more dynamic credit scores. AI is also being used to predict borrowers’ probability of default based on behavioural patterns.
The shift comes against the backdrop of a sharp increase in stressed loans. Kenya’s banking sector recorded approximately KSh717.4 billion ($5.56 billion) in gross non-performing loans in March 2025, the highest level in more than two decades. The gross NPL ratio reached 17.4%, compared with 16.4% in late 2024, before rising further to 17.6% in August 2025.
By May 2026, however, the NPL ratio had eased to around 15.3%, representing approximately KSh694.8 billion ($5.38 billion) in bad loans against a total industry loan book of about KSh4.54 trillion ($35.19 billion).
The Central Bank of Kenya estimates that nearly 45% of the country’s commercial banks are already using AI for credit risk assessment. The technology is allowing lenders to move away from traditional credit models heavily dependent on documentation and physical collateral such as property title deeds.
AI-based systems can assess factors such as transaction patterns, bill-payment behaviour, business activity and changing economic conditions to develop borrower-specific risk profiles. They can also continuously monitor customers after loans are disbursed, potentially identifying early signs of financial stress before repayment problems materialise.
According to the report, advanced systems can incorporate variables including inflation, exchange-rate movements, commodity shocks, unemployment trends and geopolitical developments. Some AI platforms claim to identify emerging repayment difficulties as much as six months before a borrower misses a payment.
This predictive capability could allow banks to shift from reactive loan recovery towards proactive credit risk management. Depending on the circumstances, lenders may be able to restructure loans earlier, initiate recovery measures or reassess exposure to higher-risk portfolios.
The Kenya Bankers Association said it remains difficult to quantify how significantly AI will ultimately reduce NPLs. However, more granular borrower assessment could enable banks to lend more accurately under the country’s Risk-Based Credit Pricing Model, where pricing reflects the risk profile of individual borrowers.
Technology providers are also targeting the Kenyan banking market with specialised credit-risk solutions. US-based IronOne Technologies LLC, in partnership with Fidem Financial LLC, is offering an AI-powered Smart Delinquency Predictor designed to forecast borrower default risks up to six months in advance.
The system continuously analyses transaction and repayment behaviour alongside external economic variables and assigns borrowers scores ranging from 0 to 999, with higher scores representing stronger credit profiles.
Rising NPLs in Kenya have been linked to several factors, including weak credit appraisal practices, high borrowing costs, deteriorating economic conditions and external shocks such as the Covid-19 pandemic and geopolitical instability. These factors have highlighted limitations in traditional models that rely heavily on historical borrower information rather than continuously changing economic conditions.
The wider adoption of AI and alternative data could also have financial inclusion implications. More sophisticated risk assessment may enable banks to evaluate individuals and small businesses with limited traditional collateral or credit histories, potentially expanding formal credit access while allowing lenders to price risk more precisely.
Want to deepen your expertise beyond today’s news?
Explore practical certification courses designed for banking, risk, insurance, compliance, ESG, AI, and emerging technologies professionals.
Learn from industry experts and earn certifications from RMAI and BFSI Sector Skill Council of India.
#Riskmanagementnews
