By Ajit Kumar, Co-Founder and Chief Operating Officer, FatakPay
Artificial Intelligence has been evolving as it becomes the defining technology in shaping financial services. Across the banking, lending and fintech ecosystem, AI is helping institutions detect fraud, assess credit, automate services and strengthen support for customers. It is also consistently working towards improving operational efficiency and compliance.
Unlike sectors where AI only enhances productivity or generates content, its services in financial services involve decision-making. AI can influence whether an individual qualifies for credit, whether a transaction is flagged as fraudulent, whether a payment is blocked, or whether additional verification is required. As AI becomes more deeply available in financial ecosystems, the institutions that generate long-term value will not necessarily be those deploying the most powerful models. They will be those deploying AI responsibly, as transparency needs to be a priority. The increasing focus on requiring trust as a basic need is also evident in the regulatory conversation.
AI is becoming the operating layer of financial services
AI is becoming the operating layer that supports, or at times impacts, everyday decisions. For example, banks use machine learning models to identify fraudulent transactions within milliseconds. AI assists in analysing spending patterns, detecting anomalies, and providing customer service through intelligent assistants. Increasingly, AI is also supporting credit assessment by enabling institutions to analyse larger volumes of structured and unstructured information more efficiently.
The pace of AI adoption is only increasing every day. Annual transaction volume expanded from just 2 crore transactions in FY 2016-17 to over 24,162 crore transactions in FY 2025-26, representing an almost 12,000 fold surge in transaction volume. Parallelly, transaction value rose sharply from ₹0.07 lakh crore in FY 2016-17 to approximately ₹314 lakh crore in FY 2025-26, translating into a more than 4,000 fold increase in transaction value.
At the same time, the National Payments Corporation of India (NPCI) has reported that UPI processed over 22 billion transactions in June 2026 alone. This demonstrated the scale at which financial institutions operate. Yet, despite AI’s growing capabilities, much of the industry conversation is still about which model delivers higher accuracy, which platform is faster or which system processes automates faster. These questions and their answers define leadership in financial services. The real question is whether AI can be trusted.
As these use cases expand, AI is becoming part of the operating layer based on which many financial decisions are made. This creates enormous opportunities where AI can reduce manual effort, boost operational speed, help prevent fraud and make financial services more accessible. However, as dependence on AI increases, there is a greater responsibility to ensure its decisions are reliable. Financial services have always operated on the principle of accountability. Every lending decision and risk assessment must ultimately be explainable, even if they are taken with the help of AI.
Black-box AI has no place in finance.
Financial decisions that remain accountable are the ones institutions can stand behind for decades. AI should be able to reinforce this principle. Customers deserve to understand why the decisions are made. Regulators expect financial institutions to justify their decisions. These automated recommendations need to be validated and survive the challenges when necessary.
This is where explainability becomes essential. AI systems cannot be a black box, whose decisions are just because the algorithm said so. Human oversight cannot be negligible here. Particularly in the case of high-impact decisions that involve creditworthiness, previous fraud, and financial risks.
Recognising these challenges, the Reserve Bank of India constituted the Committee on the framework for Responsible and Ethical Enablement of AI (FREE-AI). The committee recommends that AI adoption in financial services needs to be fair and transparent. In particular, the framework looks out for privacy and security. It argues that responsible AI is what enables innovation to scale safely and sustainably. This represents an important shift in perspective. Governance is no longer about slowing technology but accepting it while ensuring security and trust.
Trust is Fintech’s Next Competitive Advantage
Responsible AI often gets treated as a compliance checklist. That undersells what it does. AI systems that are tested regularly, built on clean data, and reviewed by people across the business, not left to the technology team alone, tend to produce outcomes that hold up under scrutiny. That matters practically: fewer errors, fewer disputes, less time spent explaining a decision after the fact.
Financial services have competed on speed for years. Faster onboarding, quicker approvals, fewer forms. But as AI capability becomes something every institution has access to, that stops being what sets anyone apart. Customers pay closer attention to how their data gets used than they did five years ago. Regulators expect institutions to account for automated decisions, not just make them. Investors are starting to ask about governance the same way they ask about growth. The institutions that take responsible AI seriously now, not because a regulator told them to but because they see where this is heading, will be the ones customers still trust in five years.