Artificial intelligence is moving beyond experimentation in banking. Large financial institutions are now shifting from isolated pilots toward enterprise-wide transformation, where AI is embedded directly into core business processes. Bank of America’s latest AI strategy reflects this transition, with the bank focusing on governance, scalability and measurable returns as it expands the use of AI across operations.
Bank of America is advancing its artificial intelligence strategy as it seeks to move beyond narrow task automation and into broader operational transformation. Speaking at the Semafor World Economy 2026 event, the bank’s Chief Technology and Information Officer Hari Gopalkrishnan outlined how the institution is evolving its approach to AI adoption across the enterprise.
The bank has already seen widespread internal and customer-facing adoption of its AI tools. Its internal virtual assistant, Erica for Employees, is now used by the vast majority of the bank’s workforce, while the customer-facing version of Erica has handled billions of client interactions since its launch. These systems have helped automate routine queries and support employees in day-to-day tasks.
However, Bank of America is now entering a new stage in its AI journey. Rather than continuing with isolated proofs of concept, the bank is increasingly focusing on using AI to transform complete business processes.
Gopalkrishnan described four priorities guiding the bank’s next phase of AI development: end-to-end process transformation, enterprise-wide scalability, governance and clear measurement of return on investment.
One of the most significant shifts involves moving AI projects from small experimental use cases into large operational workflows that can deliver measurable impact. Instead of building tools that solve individual tasks, the bank is targeting broader processes that influence customer experience, revenue generation and cost efficiency.
Data remains central to this strategy. According to Gopalkrishnan, the effectiveness of AI depends heavily on the quality and availability of structured data. Banks must also carefully manage the infrastructure required to run AI models, including compute resources and the financial cost of operating advanced models.
The economics of AI are becoming a major consideration for large institutions. Training and running advanced models requires significant computing power, making cost management and financial oversight an important component of AI strategy. Technology leaders are increasingly working with financial operations teams to monitor the cost of AI infrastructure and ensure projects deliver tangible value.
Another major focus area is scale. Rather than allowing separate teams to build isolated AI tools, Bank of America is attempting to create enterprise-level AI capabilities that can be reused across different business units and processes. With thousands of internal workflows across the bank, scalable AI platforms could potentially support automation and decision-making at a much broader level.
Governance is also becoming increasingly important as AI adoption accelerates. Financial institutions must balance innovation with risk management, particularly when deploying models that influence operational decisions or customer interactions. According to Gopalkrishnan, implementing governance frameworks that allow innovation while maintaining appropriate oversight remains one of the biggest challenges in enterprise AI deployment.
The bank is also investing heavily in workforce development to support its AI strategy. Training programmes have been introduced to help employees build AI skills ranging from basic prompt engineering to more advanced design and development capabilities. By improving AI literacy across the workforce, the bank aims to ensure employees can effectively work alongside AI systems as they become more deeply integrated into operations.
In wealth management, the bank has already begun applying AI tools to client interactions. One example is an AI-powered meeting support platform that assists financial advisers before, during and after meetings with clients. By analysing internal CRM data and other information sources, the system can help advisers identify potential prospects, prepare for meetings and automatically generate summaries afterwards.
According to the bank, this technology can significantly reduce the time required for administrative tasks and improve adviser productivity.
As generative AI investment continues to rise across the corporate sector, banks are increasingly exploring how AI can support large-scale operational transformation. Industry forecasts suggest organisations will significantly increase spending on generative AI in the coming years, further accelerating the integration of AI into core business workflows.
What this means for the industry
- Banks are shifting from AI experimentation to enterprise transformation. Early pilot projects are giving way to AI initiatives designed to reshape entire operational processes.
- Return on investment is becoming a critical measure of AI success. Financial institutions are under increasing pressure to demonstrate tangible value from large AI investments.
- Governance frameworks will determine how quickly banks can scale AI. Institutions must balance innovation with regulatory and operational risk.
- Data infrastructure remains the foundation of effective AI deployment. Without strong data management and architecture, AI projects struggle to scale.
- Workforce reskilling is becoming a central component of AI strategies. Banks are investing in employee training to ensure staff can effectively work alongside AI tools.
Photo by Javier Haro
