J.P. Morgan logo (Wikimedia Commons, public domain)
JPMorgan CFO says the bankās AI exposure remains modest as sentiment shifts
Jeremy Barnum says AI-related spending has come in below expectations, even as the bank keeps its lead in banking AI deployment
JPMorgan Chase wants you to know it is not betting the house on artificial intelligence. Chief Financial Officer Jeremy Barnum says the bank’s AI exposure is modest, though it may be affected by shifting market sentiment.
What Barnum actually said
On October 13, 2026, Barnum reported that JPMorgan’s AI-related expenses have so far run lower than the bank anticipated this year. The spending has not spiraled. If anything, it has undershot the plan.
The caveat matters, though. Barnum flagged that this modest exposure could still be influenced by changing sentiment toward AI.
In February 2026, Barnum said JPMorgan’s exposure to the software sector remains limited relative to its much larger wholesale portfolio. That comment landed amid broader worries that AI could disrupt software businesses. He noted that the bank’s software exposure is concentrated in enterprise software.
The numbers behind the bank’s AI machine
JPMorgan runs a technology budget of approximately $20 billion a year, and a substantial share of that goes toward AI initiatives.
That money powers more than 1,000 AI use cases across the bank. Approximately 150,000 employees use the bank’s internal language models every week.
AI, tech, and the markets they move—in one daily briefing.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
JPMorgan ranked first in the Evident AI Index for Banks for the fifth consecutive year as of October 2026.
Why the bank is playing it cool
JPMorgan recently released earnings that missed profit expectations. The shortfall was partly attributed to softer trading activity.
CEO Jamie Dimon has said AI is transforming how the bank operates and delivering significant productivity gains. Those gains have come with some headcount reductions in specific units.
What this means for investors and rivals
For JPMorgan shareholders, the takeaway is mostly about risk. A bank with limited software lending exposure and below-plan AI expenses is less vulnerable if AI enthusiasm cools sharply.
For rival banks, JPMorgan’s approach offers a template worth studying. The bank is pairing broad internal adoption, with 150,000 weekly users and over 1,000 use cases, against a spending pace that has actually come in lighter than expected.
The workforce angle is another thread to follow. Dimon’s acknowledgment of headcount reductions in certain units suggests the productivity benefits are real enough to show up in staffing decisions, at least in parts of the bank.