At the Portfolio Future of Finance 2026 conference, Léder Tamás, Director of Digital Business Competencies at MBH Bank, discussed how AI agents may reshape banking processes over the coming years. He said that while it is hard to predict the exact trajectory of artificial intelligence, practical, customer-facing solutions are already in use today.
Current state
Léder noted that financial institutions — banks, insurers and other providers — are increasingly deploying concrete AI solutions. He pointed out that MBH Bank already uses chat-based customer service agents that operate 24/7 to assist with matters such as payroll loan applications and general banking questions. According to him, this level of implementation is where banks should be testing the technology, even as more complex developments continue behind the scenes.
Which processes could AI take over?
He expects substantial progress within 5–10 years: processes that currently require lengthy paperwork and weeks of administration, such as mortgage applications, could be carried out with the help of AI agents. Léder also anticipates significant automation in development and operational back-office workflows, reducing manual effort and accelerating internal processing.
Who benefits and why?
Because younger generations demand digital service, Léder argues that an increasing number of banking processes will occur without human intervention. At the same time, he stressed that personal contact and human dialogue will not disappear; their value may even increase for situations where digital tools do not provide sufficient guarantees or where complex financial advice is needed.
Risks and regulatory questions
The MBH Bank executive warned that AI is not infallible: systems can make errors or “hallucinate,” so human advisers and experts will remain necessary. He also highlighted regulatory gray areas: if an AI tool supports investment or financial decision-making, it raises the question whether that activity should be classified as regulated financial advice under existing legal and supervisory frameworks.
Ethical and safety considerations
Léder addressed broader concerns about AI posing a threat to humanity, referencing Isaac Asimov’s laws of robotics. He suggested that AI systems should be constructed with human-protective logic and appropriate safeguards and regulation to prevent harm.
Conclusion
In summary, Léder Tamás believes AI will become an important tool in the financial sector: useful customer-facing applications already exist, and in the next 5–10 years more complex, paper-based and time-consuming processes are likely to be automated. However, due to the possibility of errors, outstanding regulatory questions and the need to preserve human oversight, adoption will proceed gradually and under supervision.
