Artificial intelligence (AI) has moved well beyond the experimentation stage in banking.
Today, it’s helping detect fraudulent transactions, verify customer identities, review loan applications, power chatbots, and analyze spending patterns in seconds. For financial institutions, AI is no longer a futuristic concept — it’s becoming part of everyday operations.
But while much of the conversation has focused on what AI can do, another question is starting to demand equal attention: Can banks use AI without creating new compliance risks?
As more financial institutions integrate AI into their systems, deploying the technology is only half the challenge. Making sure it complies with regulations, protects consumers, and earns public trust may prove even more difficult.
The technology is moving faster than the rules
IMAGE CREDIT: Shutterstock
AI’s ability to automate decisions and process enormous volumes of information is one of its biggest strengths. It’s also one of the reasons regulators are paying close attention.
Unlike traditional software, AI can generate responses, recommend actions, or make predictions that aren’t always easy to explain. If a chatbot gives inaccurate financial advice, an AI model introduces bias into credit decisions, or an automated system misses suspicious transactions, it’s the bank—not the technology—that remains accountable.
That’s where compliance comes in.
For banks, AI compliance isn’t simply about following a checklist. It’s about ensuring AI systems operate within legal, ethical, and regulatory boundaries while producing results that can be trusted by customers, auditors, and regulators alike.
Around the world, governments are beginning to establish guardrails for responsible AI use. The European Union’s AI Act, the General Data Protection Regulation (GDPR), emerging AI policies in the United States, and international ISO/IEC standards are among the frameworks shaping how organizations develop and deploy AI.
The common goal is straightforward: encourage innovation while reducing the risks that can come with increasingly autonomous systems.
Where banks could run into trouble
IMAGE CREDIT: Freepik
In financial services, compliance extends well beyond making sure an AI model produces accurate answers.
Every AI-assisted interaction must meet the same regulatory expectations that apply to human employees. Whether it’s screening transactions for potential money laundering, reviewing Know Your Customer (KYC) documents, or supporting lending decisions, banks remain responsible for the outcome.
That creates a different set of challenges.
An AI-powered customer service assistant, for example, might fail to provide a disclosure required before completing a transaction. An automated complaint-handling system could misunderstand what a customer is reporting and fail to record it properly. AI may also struggle to recognize situations that require human intervention, particularly when customers raise complex regulatory or financial concerns.
Cross-border banking introduces another layer of complexity. Financial institutions operating across multiple jurisdictions must navigate different privacy laws, consent requirements, disclosure rules, and emerging AI regulations—all of which continue to evolve.
These aren’t theoretical concerns. As AI becomes more deeply embedded in financial services, regulators around the world are increasingly examining how banks govern, monitor, and validate AI-powered systems.
The Philippine banking sector is facing the same challenge
The Philippines is no exception.
According to workflow automation platform Kissflow, banks are under increasing pressure to modernize compliance processes as regulations evolve and digital financial services continue to expand.
The difficulty is that many institutions are trying to introduce AI into technology environments that were built years — even decades — ago. Legacy infrastructure isn’t necessarily incompatible with AI, but integrating modern tools often requires significant upgrades in governance, data management, and risk controls.
Recognizing this shift, the Bangko Sentral ng Pilipinas (BSP) issued Memorandum No. M-2026-031 earlier this year, providing guidance for BSP-supervised financial institutions on developing AI governance and risk management frameworks.
Rather than focusing solely on the technology, the central bank is encouraging financial institutions to think about how AI will be managed throughout its lifecycle.
Its guidance is anchored on five principles summarized by the acronym STARS: Sustainability, Transparency, Accountability, Responsibility and Security.
Together, these principles reinforce an idea that’s becoming increasingly important: responsible AI doesn’t happen by accident. It requires clear governance, continuous oversight, and people who remain accountable for the technology they deploy.
AI should support human judgment — not replace it
AI is already finding its way into many parts of banking.
It helps identify unusual payment patterns, strengthens fraud detection, assists with customer verification, estimates credit risk, recommends financial products, and enables chatbots to answer routine inquiries around the clock.
Those capabilities will almost certainly continue to expand.
Even so, the BSP has consistently emphasized that AI should support human decision-making rather than replace it.
Banks are encouraged to review and validate AI-generated outputs instead of accepting them automatically, particularly when decisions could significantly affect customers’ finances or access to financial services.
That distinction matters.
Customers may interact with an AI assistant, but accountability still rests with the financial institution behind it.
Innovation and trust must go hand in hand
There’s little doubt that AI will continue reshaping banking over the next few years. The technology promises faster operations, more personalized services, and greater efficiency across almost every aspect of financial services.
But innovation alone won’t determine which institutions succeed.
As AI takes on a bigger role, banks will be judged not only by how quickly they adopt new technologies, but by how responsibly they use them. Compliance is no longer just a function of the legal or risk management team—it’s becoming part of the foundation for building customer trust in the AI era.
The technology will keep evolving. Regulations will evolve with it.
What probably won’t change is this: when an AI-powered decision affects a customer’s money, financial well-being, or trust, people — not algorithms — will still be expected to answer for it.
