BFSI firms set to see AI-led decision-making and real-time fraud detection

Banking and financial services companies foresee AI-assisted decision-making and real-time fraud prevention becoming common in the near future, as early adopters move beyond experimentation to operational deployment, according to the ET-Cisco AI Readiness & Adoption survey for the BFSI sector.

Meanwhile, the broader BFSI industry’s enterprise readiness is under the spotlight, even as companies plan to scale up AI deployment. Before large-scale adoption, companies need to address key challenges around AI governance, infrastructure and data security.

“The biggest barrier to moving from experimentation to production is not technology readiness, but enterprise readiness,” said Ashish Mittal, Chief Technology Officer at Tata AIG General Insurance. Implementation is often slowed by fragmented data, weak governance frameworks, integration challenges with legacy infrastructure. and difficulties in scaling pilots sustainably, Mittal said.


AI moves to the core
Industry experts say the BFSI sector has already moved beyond the first phase of AI adoption focused on chatbots and customer servicing. “AI in BFSI is no longer just customer-facing experimentation. It has already moved into core operations,” said Anand Mihir, Partner and Financial Services Consulting Leader (Domestic) at EY.

Mihir highlighted customer onboarding, collections, enterprise knowledge management, sales copilots, audit review, and policy compliance work, among others, where companies are deploying AI in the sector. “Fraud and anti-money laundering have used machine learning for years,” he said.

In insurance, companies are deploying AI across risk assessment, predictive analytics and claims workflows, said Girish Nayak, chief enterprise AI and technology officer at ICICI Lombard. While several of the 65 companies surveyed said they were actively scaling AI across business functions, a majority are still in the early adoption phase.

Building the foundation
Executives believe that firms successfully scaling AI are building enterprise-wide foundations rather than isolated deployments.

According to Mihir, successful firms are linking AI strategy with business outcomes, investing in unified data platforms, building regulatory guardrails and upskilling teams across business and technology functions.

Nayak said the industry is moving away from “black-box AI” towards more transparent and responsible AI systems, pushing firms to invest in stronger governance, model transparency and audit-ready systems.

Early fruits
The strongest returns from AI deployments are currently coming from customer service, collections and operational automation, Mihir said. GenAI copilots in software development and quality assurance are delivering productivity gains of 30-40% for developers, he added.

Over the next few years, BFSI firms are expected to pursue a dual-track AI strategy focused on both operational productivity and customer growth.

This article is part of the AI Vantage series, developed in partnership with Cisco.