Artificial intelligence (AI) is being adopted more widely across Hungarian banks and insurers, but experts disagree on its long-term effects on costs, profitability and regulation.
How institutions are using AI
-
K&H Bank: According to Vadócz Zsolt, AI tools now provide highly specific daily support. K&H’s Kate assistant can do more than chat; it is able to complete certain banking tasks end-to-end. Early AI projects focused on extracting information from internal documents and automating administrative tasks; more recently agents are also helping with customer service and analysing data generated at front ends.
-
MBH Bank: Léder Tamás says MBH primarily deploys AI in back-office areas such as reporting and lending processes. He noted that while internal process optimisation will lower operating costs, payments to AI providers will offset some of those savings. He also stressed the need for service providers to remain in place — if a provider disappears due to AI-driven changes, the service may vanish too.
-
Uniqa: Kurtisz Krisztián expects major changes in operations, sales and internal processes. He observed that insurtech has already advanced sales processes, but a shift to agentic (agent-based) operation could have an even larger impact on the insurance market. Today traditional insurers have a cost ratio of 15–18 percent; Kurtisz suggested an agentic model could reduce that to about 5 percent.
Profitability, costs and regulation
Experts diverge on whether AI will ultimately preserve or erode profits. Victor Tamás Máté (Hold) argued that it is mistaken to assume efficiency gains will simply lead to mass layoffs; instead, he warned, ChatGPT-like systems that do everything for customers could erode banks’ profitability. Léder said that at least some firms will retain profitability.
Regulatory uncertainty is a significant challenge, especially concerning which data may be shared with AI systems. Market participants worry that unclear rules could lead to fines in future audits.
Examples of applications and effects
- At K&H, the Kate agent can carry out full banking procedures rather than just provide information.
- MBH uses AI heavily in internal reporting, lending and other administrative processes; the next hurdle is embedding AI usage into compliance and across suppliers.
- Uniqa has piloted end-to-end claims handling with open AI models; bringing such solutions in-house could eliminate much of the administration. The insurer’s operational claims headcount is already under 20 percent of total staff, with remaining employees focused on compliance, IT security, controls and sales.
Why this matters
In the short term, AI promises significant efficiency gains and cost reductions in finance, particularly through automating administrative and repetitive tasks. However, provider costs, changing business models and regulatory ambiguity raise questions about net savings, long-term profitability and market stability.
Event note
The article also mentions a Portfolio AI & Digital Transformation conference scheduled for 26 November 2026.
(Spokespeople cited: Victor Tamás Máté — Hold, Vadócz Zsolt — K&H Bank, Léder Tamás — MBH, Kurtisz Krisztián — Uniqa.)
