Strategic Intelligence: Agentic AI in Banking

This report starts with a simple and inclusive definition of agentic AI, which captures the notion of a general evolution toward more autonomous capabilities, rather than strict qualifying criteria. It then maps out the key gridlines of change across technology, regulation, macroeconomic, and social trends, which will shape the evolution of agentic AI over the next 24 months. The report shares executive survey data to map out where tech decision-makers at banks are prioritizing investments, both across different technologies and within AI. It explores how early adopters have delivered the most business impact, reviewing organizational models around go-to-market and implementation. The report concludes with firm-level summaries of financial services providers and technology partners, summarizing their competitive position within the theme.

AI is a broad, longstanding, and dynamic category of tech disruption within financial services. For many years, machine learning algorithms were the central focus of analysis, helping refine various aspects of banking, such as credit risk, fraud detection, and product recommendations. More recently, generative AI (GenAI) has exploded into the mainstream, and that trajectory toward more autonomous tools has led to agentic AI becoming the next logical step in the overarching path toward artificial general intelligence. This is still some way off, but market participants are beginning to develop their GenAI workflows, systems, processes, and governance with a view to agentic AI being the next evolution-building in relevant capabilities and understanding so they can supercharge efforts in the coming years.

Scope

• When Celent (part of the analyst) surveyed bank executives on their top technology priorities for 2025, “artificial intelligence and advanced data analytics” was ranked first by the highest number of respondents.
• Reinforcement learning is the prominent technique enabling AI agents to make decisions based on feedback from their environment. By learning through experience, AI agents are not restricted to following explicit or pre-programmed instructions for each scenario they are faced with; they can adapt to new situations independently.
• Between 2024 and 2025, the global proportion of consumers who had fallen victim to fraud in the preceding three years rose from 18% to 28%. Agentic AI capabilities are already being leveraged in a diverse range of fraud detection and mitigation scenarios to respond to this trend.

Reasons to Buy

• The analyst’s Strategic Intelligence is a single, integrated global research platform that provides an easy-to-use framework for tracking all themes across all companies in all sectors.
• This report is essential reading for senior executives at financial services companies seeking to understand how to reassess and potentially recalibrate AI efforts in accordance with evolving market conditions.