Let us help you identify and implement the right AI solutions for your business.
As bunq experienced rapid growth as the second-largest neobank in Europe, scaling customer support became a critical challenge. With millions of users demanding personalized banking information on accounts, spending patterns, and financial advice on demand, the company faced pressure to deliver instant responses without proportionally expanding its human support teams, which would increase costs and slow operations.[1] Traditional search functions in the app were insufficient for complex, contextual queries, leading to inefficiencies and user frustration.
Additionally, ensuring data privacy and accuracy in a highly regulated fintech environment posed risks. bunq needed a solution that could handle nuanced conversations while complying with EU banking regulations, avoiding hallucinations common in early GenAI models, and integrating seamlessly without disrupting app performance.[2] The goal was to offload routine inquiries, allowing human agents to focus on high-value issues.
bunq addressed these challenges by developing Finn, a proprietary GenAI platform integrated directly into its mobile app, replacing the traditional search function with a conversational AI chatbot. After hiring over a dozen data specialists in the prior year, the team built Finn to query user-specific financial data securely, answer questions on balances, transactions, budgets, and even provide general advice while remembering conversation context across sessions.[2][3]
Launched as Europe's first AI-powered bank assistant in December 2023 following a beta, Finn evolved rapidly. By May 2024, it became fully conversational, enabling natural back-and-forth interactions. This retrieval-augmented generation (RAG) approach grounded responses in real-time user data, minimizing errors and enhancing personalization.[4]
Book a free consultation to explore how AI can solve your specific challenges.
Development of Finn began in earnest after bunq invested heavily in data talent, hiring more than 12 data specialists throughout 2023 to build robust infrastructure for GenAI integration. This groundwork addressed key challenges like secure data retrieval and model fine-tuning for financial accuracy. The beta phase concluded by late 2023, with public launch on December 19, 2023, positioning bunq as Europe's first AI-powered bank.[1][2]
Post-launch iterations were swift: initial rollout focused on basic querying of spending and accounts. By May 14, 2024, Finn achieved full conversational capabilities, supporting multi-turn dialogues and context retention. This timeline reflects bunq's agile approach, leveraging its RESTful API for seamless app integration.[4]
Finn employs a retrieval-augmented generation (RAG) architecture, combining large language models with bunq's proprietary financial data pipelines. User queries trigger secure, real-time retrieval from personal transaction histories, balances, and budgets, ensuring responses are hyper-personalized and compliant with GDPR. The system was trained on anonymized datasets to handle Dutch, English, and other EU languages, reducing latency to under 2 seconds per response.
Key implementation details include fine-tuning LLMs (likely based on models like GPT variants) on banking-specific prompts to mitigate hallucinations—critical in finance where errors could lead to compliance issues. Integration via bunq's open banking API allowed programmatic access to cards, payments, and multi-IBAN management, enhancing Finn's utility.[3][5]
Finn is embedded as the app's core interface, accessible via a chat icon, replacing clunky menus. Users can ask natural questions like "How much did I spend on coffee last month?" or "Suggest a budget for travel," with Finn visualizing data via charts. Privacy is paramount: all processing occurs on-device or via encrypted endpoints, with opt-in data sharing.
Rollout strategy involved A/B testing during beta with select users, gathering feedback via in-app surveys to refine accuracy from ~70% to over 90%. Challenges like handling edge cases (e.g., fraud disputes) were overcome by escalating to humans seamlessly, with Finn summarizing context for agents—reducing handover time by 50%.[6]
Initial hurdles included ensuring regulatory compliance and model reliability. bunq's data team implemented guardrails, such as response validation against transaction logs, and continuous monitoring for bias. Cost challenges were mitigated by optimizing inference on efficient cloud infrastructure, keeping per-query costs low. User adoption was boosted through app updates and live demos, like Update 24 streams.[5]
By 2025, Finn supports advanced features like predictive analytics, with ongoing expansions planned per bunq's Update 30 announcements. This phased implementation—beta, launch, conversational upgrade—demonstrates scalable GenAI deployment in fintech.
Discover how we can help you implement similar solutions.
Founder & Partner
Reruption GmbH
Falkertstraße 2
70176 Stuttgart
Phone