Global banks have made significant AI investments, yet only 40% consider themselves AI leaders and even fewer effectively manage customer trust or AI reliability, according to EY's 2025 AI Confidence Pulse research. As boards press for tangible returns, a new EY analysis outlines five attributes that separate AI leaders from laggards.
The survey of 50 global banks found that eight in ten have seen efficiency and productivity gains from early AI deployments. However, just 53% reported revenue growth, and a small group of top performers is pulling ahead. The latest Evident AI Banking Index shows that leading banks are improving more than twice as fast year over year as the rest of the market.
"The challenge isn't a lack of AI use cases, but narrowing the options to those that generate real value, measured against financial returns and strategic goals," said Beatriz Sanz Sáiz, EY Global AI Sector Leader.
What effective AI strategies look like
EY's framework identifies five core attributes. First, a bold and coordinated vision that avoids fragmented pilots. Sameer Gupta, EY Global Financial Services AI Leader, said, "Without a platform-based approach, banks risk creating 15 versions of the same AI function - just like they once used 15,000 spreadsheets." Centralized infrastructure and reusable AI agent libraries help maintain consistent logic across use cases, from fraud detection to stress testing.
Second, strong governance is essential. The EY October 2025 Responsible AI Pulse Survey found that banks with formal AI oversight committees and real-time monitoring are significantly more likely to achieve revenue growth and cost savings. A governance watchtower approach - combining automated controls testing, human oversight, and continuous model validation - has emerged as a leading practice. Standardization across data security, vendor management, and model governance enables responsible scaling.
Third, AI must be business-led and focused on strategic issues, not treated as a technology backlog. Preetham Peddanagari, EY UK&I Chief Technology Officer, said, "AI isn't a backlog of use cases; it's a capital allocation decision. Fund the few bets you can govern and measure end to end, and shut down everything that can't prove both value and trust at scale." Leading banks are using digital twins for treasury cash flow simulations and automating the full order-to-cash lifecycle in commercial banking.
Fourth, human-centered change management often determines success. "The biggest barrier to scaling AI isn't the algorithm - it's the change management," Gupta said. "Training teams, reworking processes and putting in the right governance often takes twice the effort of building the model itself." Banks must design human-in-the-loop workflows for tasks like credit decision validation and create new roles such as prompt engineers and AI workflow designers. An EY/MIT report found that more than three-quarters of executives now view agentic AI as a coworker, fundamentally shifting how workflows and governance models should be built.
Finally, long-term planning means designing modular, scalable infrastructure and preparing for convergence with digital assets, tokenization, and quantum computing. Banks are exploring hybrid cloud and on-premises large language models for sensitive use cases to balance risk and innovation.
Recommended actions for C-suites and boards
To move beyond pilots, EY recommends several immediate priorities:
- Set a clear vision from the top, directly linked to strategic goals and financial targets, and define enterprise data standards.
- Establish an ROI-based roadmap tying AI investments to EBITDA and P&L impact, alongside business metrics like risk reduction and customer engagement.
- Create a governance watchtower with board-level oversight, multilayered controls, and automated monitoring for the entire AI lifecycle.
- Prioritize data lineage and quality by deploying AI tools to track data used in large language models, including third-party data.
- Shift accountability from IT to business leaders for market-facing processes, and encourage long-term thinking about AI's role in product development and client service.
- Engage regulators proactively to help shape standards for data security and ethical AI usage.
- Invest in upskilling and culture change, enabling middle management to foster AI-friendly experimentation and addressing the 56% of desk-based employees who worry about job security despite enthusiasm for AI agents.
Why this matters for executives and strategy
For banking executives, the gap between AI investment and returns is a board-level concern. The path to outperformance requires shifting from IT-led experimentation to business-owned, governed, and human-centered AI programs. Leaders who can articulate a bold vision, enforce disciplined governance, and drive cultural change will be positioned to capture the next wave of value - while those who treat AI as a technology project risk falling further behind. Many strategy leaders are finding practical frameworks through resources on AI for Executives & Strategy to bridge this execution gap.
Your membership also unlocks:
