This project was implemented through the 'Public-Private Secondments for SupTech Innovation' initiative between the Central Bank of Kenya and DANA Indonesia.
The Public-Private Secondments for SupTech Innovation initiative pairs highly trained experts from the private sector with financial authorities around the world for short-term, remote and in-person secondments that accelerate the digital transformation of financial supervision, leading to a more efficient, safer and sounder global financial system.
The Central Bank of Kenya is advancing its data capabilities through the Granular Data Integration (GDI) project, aimed at centralizing structured supervisory data for improved oversight. A critical next step is to explore whether this data can power machine learning models to forecast liquidity and credit risk at the institutional level.
Currently, CBK relies partly on institution-reported stress tests. However, CBK seeks to strengthen its internal modeling capabilities by using its own data to simulate risk scenarios and proactively identify emerging vulnerabilities. This secondment project will assess whether the anonymized GDI data can support supervised ML approaches for risk prediction. If so, the project team will proceed to design and prototype a basic AI/ML model, focusing on data preparation, feature identification, and algorithm selection. If the data proves insufficient, the team will identify and propose another viable use
case aligned with CBK’s supervisory priorities. This initiative represents a foundational step in embedding AI/ML into CBK’s supervisory processes.
