Financial Institutions Shift to Multi-Agent AI to Drive Bottom-Line Value

Financial institutions are shifting from isolated AI experiments to sophisticated multi-agent systems to handle core banking and insurance workflows. While adoption is accelerating, a new report highlights that scaling these autonomous systems requires a significant overhaul of underlying cloud infrastructure and a disciplined focus on measurable outcomes rather than mere efficiency gains.

What was announced

The Capgemini Research Institute’s World Cloud Report for Financial Services 2027 indicates a major transition in how banks and insurers approach artificial intelligence. Currently, 59% of firms are either piloting or deploying multi-agent AI in production environments. However, achieving true autonomy remains a challenge, with only one in ten organizations having reached the level of orchestration necessary to run these agents autonomously across the entire enterprise.

To bridge this gap, 80% of executives are prioritizing the reduction of technical debt as a prerequisite for success in the agentic AI era. The criteria for funding these initiatives are also shifting rapidly. By 2028, the number of executives prioritizing specific outcomes as their primary funding metric is projected to rise by 141%, while those focused solely on efficiency are expected to drop by 42%. This data comes from a global survey of 1,100 C-suite executives of financial services firms.

The research also found that approximately half of financial services firms struggle to demonstrate a consistent return on investment for agentic AI initiatives and lack clear frameworks for assessment. In response, many are turning to intelligent model routing to direct tasks to the most cost-effective and accurate models. Furthermore, over half of the industry identifies hyperscale cloud providers as preferred partners for governing these deployments, citing the need for regulatory compliance, data residency certifications, and robust security and identity frameworks at scale.

The companies involved

Capgemini is a global leader in consulting, technology services, and digital transformation. The firm operates at the intersection of business and technology, providing a broad range of services including cloud, data, and artificial intelligence strategy. With a significant presence in the financial services sector, the company frequently publishes research through the Capgemini Research Institute, offering insights into how banks, insurers, and capital markets firms are adapting to emerging technologies.

The organization has established itself as a key partner for financial institutions looking to modernize their legacy systems and migrate to cloud-native architectures. Ravi Khokhar serves as the Executive Vice President & Head of Cloud at Capgemini, overseeing the firm's cloud strategy for the financial sector. The company's focus on "agentic AI" reflects a broader market trend where firms are moving away from simple chatbots toward complex, interconnected AI systems capable of executing multi-step business processes with minimal human intervention. This shift is driving a hybrid sourcing model where firms balance in-house development with packaged solutions and ecosystem partnerships.

What FF News has reported before

FF News has extensively covered Capgemini’s influence across the financial landscape, including its role as a strategic technology partner for the UK’s Financial Services Compensation Scheme. The firm’s research often highlights systemic risks and opportunities, such as the revenue risks banks face from stablecoins and tokenized deposits disrupting global payments.

In the insurance sector, FF News reported on Capgemini data showing how life insurance relevance drives higher revenue growth, as well as the company’s recognition as an AI pioneer at the Guidewire Connections event. These reports collectively underscore Capgemini's position as a primary source of data-driven insights for the fintech and insurtech industries.

What this means

The shift toward multi-agent AI signals the end of the "experimentation phase" for generative AI in finance. The industry is moving toward a reality where AI is not just a tool for answering queries but an autonomous workforce. This puts immense pressure on legacy technology stacks; banks that fail to resolve their technical debt will find it impossible to orchestrate these agents effectively. The move toward hyperscalers also suggests a consolidation of power, as smaller cloud providers may struggle to meet the rigorous regulatory and security demands of autonomous finance. The central question for the sector is no longer whether AI works, but whether the existing infrastructure can handle its autonomy without creating new systemic risks.