By Gisela Colombo
In normal times, artificial intelligence is typically presented as a tool to improve efficiency, personalize customer experience, or reduce operating costs. In Ukraine, however, the conversation carries different weight. There, financial technology is measured not solely by its capacity to innovate, but by its ability to sustain critical services under extreme conditions.
War transformed Ukraine’s financial system into an involuntary laboratory for digital resilience. Banks, fintechs, regulators, and technology providers had to answer an urgent question: How does banking remain functional when physical infrastructure can be attacked, branches become insufficient, and millions of people depend on digital channels to withdraw, pay, transfer, or access their funds?
In this context, artificial intelligence does not appear as a decorative layer over traditional processes. It emerges as a practical tool to reduce friction, automate tasks, detect risks, and sustain operations. Its value lies not in replacing banking, but in making it more resilient.
One of the most visible examples is Monobank, one of Ukraine’s most recognized digital banks. The institution incorporated AI to simplify payments by IBAN: users can upload an invoice or invoice in PDF, PNG, or JPEG format, and the system automatically identifies the necessary data to complete the payment. It seems like a minor improvement in user experience, but in a country at war it carries deeper meaning. Every manual step eliminated reduces errors, accelerates operations, and facilitates access to financial services at moments when attention, time, and connectivity may be limited.
PrivatBank offers another dimension of the same phenomenon. At the start of the Russian invasion, the bank migrated critical applications to the cloud on an extremely short timeline to preserve service continuity. This decision is not AI per se, but rather the foundation upon which AI can operate: flexible infrastructure, accessible data, distributed systems, and crisis response capability. Without that architecture, any ambition for artificial intelligence remains trapped in fragile systems.
What is interesting about the Ukrainian case is that it combines three layers: accelerated digitalization, operational necessity, and regulatory oversight. The National Bank of Ukraine already recognizes that AI and machine learning are being used by a significant portion of the financial market, and at the same time works on guidelines to ensure that adoption is responsible. The concern is not only to innovate, but to avoid new risks: opaque automated decisions, bias, privacy issues, cyberattacks, technological dependency, or failures that could affect trust in the system.
This tension is key. In countries at war, AI can be a competitive advantage, but it can also become a vulnerability if implemented without governance, without traceability, or without human oversight. A model that automates credit decisions, identifies suspicious transactions, or responds to customers must be efficient, but also explainable, auditable, and secure. Resilience does not consist solely in continuing to function; it consists in continuing to function without breaking trust.
Ukraine demonstrates that the banking of the future does not necessarily emerge from the most comfortable markets, but from those forced to solve urgent problems. There, AI becomes an answer to very concrete questions: how to operate with less physical dependence, how to better serve customers under pressure, how to detect anomalies faster, how to automate documentation, how to sustain payments, and how to protect data in a hostile environment.
For Latin America, the lesson is clear. There is no need to wait for a war to think of banking as critical infrastructure. Economic crises, natural disasters, cyberattacks, political instability, or connectivity problems also demand more adaptive financial systems. The question is not whether banking will incorporate AI, but for what purpose it will do so.
The Ukrainian case leaves a powerful lesson: artificial intelligence in banking should not be thought of solely as a race to become more efficient. In the most difficult contexts, its true value lies in helping the financial system remain standing.
