Monica Hovsepian is the Global Financial Services Lead at OpenText.
Banks spent the last decade automating tasks. The next phase is different. AI is starting to make decisions, coordinate workflows and take action across systems with limited human intervention.
This shift is known as agentic AI. Unlike traditional automation, agentic AI does more than respond to prompts. It interprets goals, accesses systems, evaluates data and initiates next steps.
For financial institutions, the implications are enormous.
According to McKinsey, generative AI and related technologies have the potential to generate between $200 billion and $340 billion annually for the banking sector through productivity gains, revenue growth and improved risk management.
But most institutions are nowhere near ready to operate at this level.
Operationalizing AI
In my work across banking and financial services, I continue to see the same pattern. Firms invest aggressively in AI pilots while avoiding the harder operational work underneath. Data remains fragmented. Governance remains inconsistent. Ownership remains unclear.
The technology is moving faster than the institution around it.
A recent Digital Banking Report sponsored by OpenText, “Agentic AI: Powering the Self-Driving Bank,” found that 96% of financial institutions are engaged in some form of agentic AI activity, yet only 19% have moved into production deployment.
That gap matters, because the competitive advantage no longer comes from experimenting with AI. It comes from operationalizing it.
Fraud detection, compliance and operational efficiency remain leading use cases because the data is structured and the business outcomes are measurable.
Institutional Data
Deloitte recently noted that AI creates the most value when workflows are redesigned rather than layered onto legacy processes.
The Digital Banking Report found that only 9% of institutions believe they can effectively connect content, communication and transactional data in real time.
According to the same research, 52% of institutions lack confidence in whether their governance systems are prepared to support compliant agentic AI deployment.
Bank of America’s Erica platform has surpassed 3 billion client interactions and handles roughly 2 million customer interactions per day.
JPMorgan Chase has deployed its internal LLM Suite platform across approximately 250,000 employees to support research, customer service and advisory workflows.
Accenture estimates that AI has the potential to increase banking productivity by up to 30% if institutions redesign workflows around the technology instead of treating AI as an overlay.
Workforce Readiness
One issue receives far less attention than it should: workforce readiness.
Many financial institutions continue to approach AI as a technology initiative instead of an operational shift. But agentic AI changes how decisions are prepared, how teams collaborate and how customer relationships are managed. Employees are no longer simply operating systems. Increasingly, they are supervising, validating and guiding AI-driven workflows.
This transition requires new skills across the organization. Relationship managers need confidence interpreting AI-driven recommendations. Risk teams need visibility into how models behave. Operations leaders need stronger oversight across automated workflows and data flows.
The Digital Banking Report found that employee enablement ranked last among institutional AI priorities at only 14%. That gap creates long-term risk. Institutions that fail to prepare their workforce for AI adoption will struggle to scale it responsibly.
The banks seeing the strongest results are treating AI transformation as both a technology strategy and a people strategy.
Challenges And Considerations
The opportunity is significant, but so are the risks. Financial institutions operate in one of the most regulated industries in the world. Questions around data privacy, model explainability, cybersecurity and accountability remain front and center.
As agentic AI takes on more responsibility, leaders must balance innovation with oversight. Moving fast matters. Maintaining trust matters more.
First Steps For Banks
Most institutions do not need another pilot. They need a clear starting point.
Begin with a business problem that has measurable value and accessible data. Establish governance early. Invest in data quality before expanding use cases. And most importantly, align business, technology and risk teams from the outset.
The banks making the most progress are not trying to transform everything at once. They are building momentum one successful deployment at a time.
Rebuilding The Institution
The conversation around AI in financial services often centers on capability. The harder question is operational readiness. The future of banking will not be defined by who adopts AI first, but by who rebuilds the institution around it fastest.
Forbes Finance Council is an invitation-only organization for executives in successful accounting, financial planning and wealth management firms. Do I qualify?
