Kenyan banks are accelerating their investment in artificial intelligence (AI) and machine learning, with deployment now concentrated in credit decisioning, fraud detection, cybersecurity, and customer-service automation.
These functions are becoming the first testing ground for AI as lenders pursue faster processing, tighter risk controls, and leaner operations.
One in two financial institutions, including commercial banks, microfinance lenders, and digital credit providers, have already integrated AI tools into at least one business process, according to a March 2025 survey by the Central Bank of Kenya (CBK).
Credit modelling has become the main use case for AI across major lenders so far, as banks strive to increase speed while maintaining prudent risk controls.
“We have, over the years, adopted AI… it is currently deployed in some critical business areas driving credit decisioning,” Dennis Volemi, the Group Director of Technology for KCB, told the Business Daily.
KCB is Kenya’s largest lender by assets. As per the CBK survey, credit has been the leading AI application across the sector, with banks using models to analyse payment histories, assess micro-borrower patterns, and automate routine lending decisions like determining whether to approve loan applications and on what terms.
At Absa, Chief Data Officer for Africa Hartnell Ndungi said AI supports credit analytics, customer lifecycle modelling, and risk monitoring, enabling more granular scoring and early detection of distressed accounts.
AI has also become a key tool in protecting banks from increasingly complex fraud schemes and cyber threats.
KCB says its fraud-monitoring systems already rely on AI to flag suspicious activity in real time, while Absa uses machine-learning models in risk-monitoring workflows.
Machine learning (ML) is a subfield of AI that uses algorithms to enable computers to learn from data, identify patterns, and make predictions or decisions with minimal human intervention.
CBK’s report shows that banks are using anomaly-detection models to identify unusual network activity, automate threat triage, and speed up incident response.
“The top three applications of AI and ML by institutions that had adopted AI were credit risk assessment at 65 percent, cybersecurity at 54 percent, and customer service at 43 percent.
This was followed by e-KYC at 41 percent and fraud risk management at 40 percent,” CBK said.
At the same time, conversational artificial intelligence, a type of AI that allows computers to understand and engage in human-like conversations using natural language, seems to be among the most mature AI applications in Kenyan banking. This includes technologies such as chatbots or virtual agents that users can talk to.
Banks are also deploying chatbots and voice-enabled interfaces to ease pressure on call centres and to serve the rising volume of digital customers.
Absa’s bank-wide conversational AI system ‘Abby’ enables customers to perform tasks such as checking balances and transferring funds through WhatsApp.
Equity Bank has a similar chatbot called ‘Eva’, available on WhatsApp, Facebook Messenger, and Telegram.
“We have deployed systems that allow natural-language querying of data, retrieval of information from policy and operational documents, and voice-enabled interfaces that support hands-free interaction and voice of customer analytics,” Absa’s Mr Ndungi said.
Another growing frontier for Kenyan lenders is the application of AI to document processing and data structuring.
Absa said its in-house platform, the Citrus AI Suite, includes systems for document intelligence, voice transcription and classification, enabling the bank to extract insights from previously unstructured data such as customer conversations, scanned documents and emails.
These capabilities support credit, risk, and service functions that traditionally relied on manual review.
These widely adopted use cases in Kenya mirror global banking trends, with credit risk assessment, cybersecurity threat monitoring, and customer engagement tools leading uptake.
Yet despite growing momentum, banks cite industry-wide constraints like skills shortages and high implementation costs as the top challenges that continue to limit full-scale adoption.
“There is high talent scarcity, making highly skilled AI and machine learning operations professionals difficult to recruit and retain,” said Mr Ndungi.
Legacy systems remain another bottleneck, as integrating advanced models into older core-banking platforms adds cost and complexity and often slows the move from pilot to production.
Per the CBK survey, for instance, 44 percent of AI adopters admit they cannot adequately explain how their models work, compounded by the fact that many institutions rely on third-party vendors.
To address these gaps, banks say they are investing in workforce capability. KCB said it is rolling out group-wide AI literacy programmes, while Absa is embedding AI into staff-facing tools to democratise analytics and operational insights.
Industry players say the business case for AI is strengthening; faster credit decisions, lower fraud losses, more accurate risk insights, and leaner operations, and that the next frontier will be overcoming skills gaps and compliance challenges to scale the systems responsibly.
“Early deployments are showing measurable impact across selected business areas,” said Mr Ndungi.
