The rapid adoption of generative AI in banking is driving significant changes in how financial institutions engage with customers and optimize their operations. KPMG’s 2025 Banking Survey reveals that nearly half of U.S. bank executives anticipate that generative AI will handle up to 40% of daily tasks by the end of the year. However, what is going on in the real world presents a contrasting narrative, with companies like Klarna adjusting their AI strategies after facing limitations in customer service automation.
Banking’s AI Push: Fraud Prevention and Personalization at the Forefront
According to KPMG, banks are investing heavily in generative AI, with 91% of executives ranking data-driven insights and personalization as key priorities. Additionally, 78% are piloting AI for security and fraud detection, underscoring the technology’s promise in streamlining operations and enhancing customer experiences. By the end of 2025, many banks expect AI to manage significant portions of their daily operations, including fraud prevention, cybersecurity, and financial forecasting.
Despite the progress, executives acknowledge that while AI is beneficial for enhancing efficiency and securing operations, releasing customer-facing AI solutions remains cautious due to concerns over data privacy and regulatory compliance. Banks are confident in AI’s long-term potential but are still refining its application within the highly regulated financial landscape.
Klarna’s AI Journey: A Cautionary Tale
Klarna, the fintech company known for its buy-now, pay-later services, offers a sharp contrast to the banking sector’s AI strategy. Initially, Klarna embraced AI as a means of reducing costs, even going so far as to replace a significant portion of its workforce with AI-powered chatbots. By 2024, the company had reduced its employee count by 40% and claimed that AI chatbots handled two-thirds of customer service conversations. However, the strategy soon backfired as the quality of customer service suffered.
CEO Sebastian Siemiatkowski recently admitted that Klarna’s heavy reliance on AI for customer service led to poor user experiences, with the chatbot solutions failing to meet customer expectations. He emphasized the importance of human support in providing high-quality service, leading Klarna to rethink its strategy. The company now plans to hire more human agents and provide customers with the option to always speak to a human when needed.
Klarna’s pivot serves as a reminder that while AI holds enormous potential, its integration into customer-facing roles can be fraught with challenges. The company’s decision to go back to human-powered customer service reflects the limitations of AI when it comes to personalization and complex customer interactions.
Balancing AI Innovation with Customer Experience
The contrast between the banking sector’s cautious optimism about AI and Klarna’s shift toward human support highlights the complexities of using AI in consumer-facing roles. Banks, while also embracing AI, appear to be taking a more gradual approach, particularly in customer interactions. They are leveraging AI for back-end operations, such as fraud detection and financial forecasting, but are hesitant to roll out AI in areas where it directly impacts customer satisfaction.
