Over 90% CXOs Exploring AI Solutions, But Most Still In Pilot Phase: Report

A growing majority of Indian financial services leaders are actively exploring artificial intelligence, with over 90% of CXOs reporting some level of AI experimentation. Yet, most projects remain in the pilot phase, underscoring both the promise and persistent challenges in scaling AI across the sector, according to the BCG + GFF Convergence AI Report.

The report finds that 56% of leaders cite an average timeline of three to six months for AI projects to move from ideation to pilot, reflecting a cautious and problem-first approach to deployment.

"We avoid any 'AI-for-cool-factor' deployments because AI use cases take at least three to six months of ideation and experimentation. For us, the approach is very clear — create a solution only if there is a real business problem, run the POC, and scale only if value is proven," said the CEO of an Indian fintech firm.

Despite growing interest, institutions face a steep climb in executing AI at scale. Talent shortages, poor data quality, and legacy infrastructure are key roadblocks that persist even after a successful proof of concept.

"Our biggest challenge is finding and developing people who can run these solutions at scale. At the same time, our data quality issues, and legacy infrastructure create additional bottlenecks, making scaled execution hard even when a POC shows clear promise," said another fintech CEO, as per the report.

According to the report, the top challenges in deploying AI solutions for financial institutions in India reflect a complex mix of capability gaps and structural constraints. Talent and skill shortages, along with data and infrastructure readiness, emerged as the most pressing issues, each cited by 57% of respondents as a top-three barrier.

These are followed by difficulties in defining high-impact use cases (45%), highlighting the challenge of aligning AI initiatives with real business value. Additionally, regulatory uncertainty (33%) and unclear returns on investment (24%) continue to weigh on decision-making. Other concerns include technology limitations (19%) and securing stakeholder buy-in (17%), suggesting that both technical and organizational readiness need to evolve in parallel for AI to move from experimentation to scalable execution.

Some leaders suggest India must think bigger: advocating for a "leapfrog" innovation strategy, akin to the success of UPI, to unlock AI’s full economic and societal potential.

"India should adopt a leap-frog innovation strategy, like UPI, to unlock the full potential of AI. Realizing this vision would need significant capital and investments in the substrate layer of AI," said a fintech founder.

As AI continues to evolve at a breakneck pace, the report stresses that manual, traditional governance mechanisms are no longer adequate. Financial institutions will need new frameworks that prioritise automation, safety, and observability.

"With AI evolving rapidly, manual oversight is no longer enough. Financial institutions must embrace new governance approaches that embed safety, observability, and automated action, while enabling structural safeguards like centralized fraud surveillance to ensure responsible adoption," said G Padmanabhan, former Executive Director of the Reserve Bank of India and Advisory Board Member, GFF.