Financial institutions are actively moving key customer-facing processes to AI agents, marking a rapid transformation in how customers interact with banks and insurers. According to the Capgemini Research Institute’s World Cloud Report in Financial Services 2026, the top processes for banks to deploy cloud-native, AI agents at scale include customer service (75%), fraud detection (64%), loan processing (61%) and customer onboarding (59%). Insurers display a similar pattern, with customer service also leading the way (70%), followed by underwriting (68%), claims processing (65%) and onboarding (59%) – collectively redefining the concept of what it means to be a financial services customer.
Recent data from the Capgemini Research Institute shows that AI agents could deliver up to US$450 billion in economic value by 2028, signalling the opportunity that exists for the financial services industry. To capitalise on this opportunity, 33% of banks say they are developing their own AI agents in-house, while 48% of financial institutions are creating new roles for employees to supervise agents.
With AI agents capable of autonomously managing complex workflows, the role of cloud is shifting beyond just an infrastructure or storage provider. Nearly two in three executives (61%) now identify cloud-based orchestration as critical to their AI strategy, transforming cloud platforms into innovation engines that can operationalise technologies at speed and scale.
“The combination of AI and cloud allows banks and insurers to tap the power of AI agents to better serve their customers with greater precision, speed and impact,” said Ravi Khokhar, Global Head of Cloud for Financial Services at Capgemini. “Our data reveals widespread industry optimism that the agentic era will open doors to new markets, signalling a new phase of transformation is upon us. To realise this potential, financial institutions must take a long-term view as humans work alongside agents. This means separating substance from hype. Leaders will need to consider how they can scale their Agentic AI operation overtime, with a vision of what they want their businesses to look like at the end of this process.”
Only 10% of firms have implemented AI agents at scale
AI agent adoption is poised for rapid growth as 80% of financial services firms are in the ideation or pilot stage of deployment. However, a sizable opportunity remains to be unlocked as only 10% of firms surveyed have implemented AI agents at scale.
According to the report, banking and insurance executives identify customer onboarding and Know-Your-Customer (KYC), processing loans and claims and underwriting as the most inefficient business functions across the sector. Firms are optimistic that Agentic AI can address these longstanding challenges, with real-time decision-making (96%), improved accuracy (91%) and faster turnaround times (89%) cited as key benefits.
Beyond efficiencies, executives view AI agents as capable of delivering real business outcomes:
· Ninety-two percent say that AI agents will help them expand into new geographies without heavy upfront infrastructure
· Seventy-nine percent believe cloud-native AI agents can unlock dynamic pricing and offers, maximising revenue and outmanoeuvring competitors
· Seventy-five percent see an opportunity for delivering multilingual support that adapts to local regulations and cultural norms
With immense potential to become a force multiplier across the enterprise, C-suite executives are aligning their investments accordingly: nearly two in three leaders indicate that up to 40% of their organisation’s Generative AI budget today is allocated specifically toward agent technologies. By 2028, one in four firms expect to increase their spending on AI agent solutions by up to 60%.
Banks face lingering challenges to adoption
As financial institutions ramp up adoption of Generative AI and AI agents, nearly all executives unanimously point to two critical roadblocks: a looming skills gap among business leaders and employees (92%) and a regulatory and compliance burden (96%). While expressing concern about the complexity of managing region-specific regulatory mandates, most executives (89%) simultaneously place compliance at the top of their organisational priorities over the next three years.
High implementation costs also emerge as a barrier to delivering real returns on AI investment. A growing number of firms (25%) are leaning towards the Service-as-a-Software model in the next 12 to 18 months, orchestrating a new approach for how AI is consumed and monetised. Rather than paying for licences and infrastructure, firms will pay for outcomes such as fraud cases resolved, transactions processed or customer queries handled.
