JPMorgan Opens Its AI Playbook to Clients

JPMorgan Chase has launched a new unit designed to give select corporate clients access to how the bank operates internally, beyond traditional merger, financing, and capital markets advice. The group, called Special Advisory Services, is intended to share internal expertise across areas including artificial intelligence, cybersecurity, digital assets, geopolitics, healthcare, supply chains, and sustainability, the bank said.

The service is available to a limited set of long-term clients, including companies preparing for initial public offerings, pursuing large transactions, or seeking to deepen their relationship with the bank. JPMorgan said it does not plan to charge for the service initially, though fees could be discussed if advisory work becomes ongoing or resource-intensive.

Executives said the initiative grew out of recurring client questions about how JPMorgan handles internal challenges. Bankers told CNBC that clients were asking for details on how the bank uses artificial intelligence, how those systems are governed, and how they are integrated into daily operations.

AI Inside JPMorgan Has Moved Past Experimentation

By 2025, JPMorgan had already integrated AI into core operations across the firm. CEO Jamie Dimon has said the bank spends roughly $2 billion annually on AI and data initiatives and has identified about $2 billion in direct benefits from those investments, including cost savings and efficiency gains.

JPMorgan has said AI tools are used across risk management, compliance, fraud detection, legal document review, and software development, rather than confined to pilot programs. Internal generative AI tools are available to tens of thousands of employees and are embedded into existing workflows.

In technology teams, the bank has reported that internally developed coding assistants improved software engineer productivity by as much as 10% to 20%, according to comments from executives.

JPMorgan has said most AI systems are designed to support human decision-making rather than replace it, with final judgments remaining with employees.

Jamie Dimon has described the financial impact as just the beginning. “We know about $2 billion of actual cost savings,” Dimon said. “I think it’s the tip of the iceberg.”

AI adoption has also influenced staffing decisions. JPMorgan has slowed hiring in some areas and allowed natural attrition to absorb productivity gains tied to automation, rather than pursuing broad layoffs, executives said.

When Clients Ask for the Playbook

As AI use expanded internally, JPMorgan executives said client conversations shifted. Rather than asking which AI tools the bank used, companies began asking how those systems were governed, how risks were evaluated, and how AI was deployed inside a regulated institution.

While other large U.S. banks have also invested heavily in artificial intelligence, none have announced a comparable effort to formalize access to their internal AI practices for clients.

Goldman Sachs has disclosed internal use of generative AI tools to assist bankers and developers but has focused those systems on productivity rather than external advisory access. Morgan Stanley has deployed AI assistants for its wealth advisers and research workflows, built in partnership with OpenAI, but has positioned those tools primarily as client-facing enhancements rather than a source of operational advisory. Bank of America has said its AI efforts are concentrated on internal automation, customer service, and risk management, including its Erica virtual assistant, without extending those internal practices into a dedicated advisory offering for corporate clients.

Special Advisory Services was created to address those requests. The unit connects clients to internal JPMorgan specialists whose primary responsibilities remain running the bank’s own operations. More than two-thirds of those experts are internally focused, limiting how many clients can be supported at a given time.

Liz Myers, who leads the initiative, compared the internal expertise being shared with traditional advisory firms. “These capabilities are on par with, or in many cases better than, what you’d find at specialized consulting firms,” she said.

JPMorgan has said the service is intended to share internal practices rather than prescribe outcomes. Clients remain responsible for implementation decisions, including AI governance and operational changes.

The material shared with clients reflects systems already in use inside the bank, including how AI deployments are reviewed, monitored, and evaluated once they move into production.

Lori Beer described the standard JPMorgan applies internally in early 2025. “Success is transforming and creating value with AI, not just incremental steps.”