JPMorgan Chase is preparing to roll out a new generation of artificial intelligence agents later this year, as the biggest U.S. bank by assets deepens its use of automation.
The systems are designed to handle more complicated assignments than earlier AI tools, which were typically used for individual requests such as drafting text, reviewing documents, or answering questions. Newer agents can move between applications, complete several stages of work, and operate for longer periods before an employee needs to step in.
Derek Waldron, JPMorgan’s chief analytics officer, said the shift marks an important step in workplace AI.
“We’ve entered now the era of long-running autonomous agents,” Waldron said. “That means that agents don’t just run for two or three minutes to carry out a goal or some instructions of a human, they can run for an hour or two.”
The bank expects to introduce these systems in 2026, although wider adoption will depend on whether companies can address security, compliance, and oversight concerns. Those issues are particularly important for financial institutions that handle sensitive customer information.
JPMorgan has already used AI to support revenue-generating teams. In private banking, its systems review market activity, client portfolios, and research overnight, helping advisers spend more time speaking with customers. Waldron said the tools have contributed to a 20% rise in gross sales and could eventually allow bankers to manage up to 50% more client relationships.
The technology is also influencing how the bank evaluates software suppliers. As AI makes it easier to build some capabilities internally, companies may become less reliant on outside vendors.
“The moat around certain types of software companies is most certainly diminished versus where it was in the past,” Waldron said.
Chief executive Jamie Dimon has acknowledged that some roles will be affected as automation becomes more capable, with JPMorgan planning to retrain and redeploy workers where possible. Waldron said companies should not treat AI solely as a way to cut costs.
“For enterprises to win with AI, it’s not about cutting the maximum number of jobs,” he said. “It’s all about trying to create a sustainable competitive advantage.”