Dimon: AI Will Hit Every Function at JPMorgan

Jamie Dimon's annual shareholder letter dropped this week and the headline writes itself: the JPMorgan Chase CEO called artificial intelligence 'transformational' — then immediately said he hesitates to use that word. Make up your mind, Jamie. What the letter actually does, buried beneath the careful language, is signal that the largest U.S. bank is betting massive on AI whether the rest of us are ready or not.

What Dimon Actually Said — and What He Didn't

The letter doesn't soft-pedal it. 'AI will affect virtually every function, application, and process in the company,' Dimon wrote in his Jamie Dimon annual shareholder letter. Every function. That's not a hedge — that's a full-throated declaration from the CEO of a bank with over 300,000 employees.

His argument on speed is the part worth dwelling on. Electricity took decades. The internet took decades. AI, Dimon says, 'looks likely to accelerate over the next few years.' If he's right — and the pace of model development over the last two years suggests he might be — then the usual buffer industries have to adapt won't exist this time. Companies won't have a decade to figure out what to automate. They'll have a few quarters.

Dimon also went full optimist on the macro effects: curing cancers, creating new composites, reducing accidental deaths. Sweeping claims. But they're not outlandish — drug discovery and materials science are already seeing real AI-driven gains. The question is whether those gains land broadly or concentrate at the top of the capital stack.

The importance of AI is real, and while I hesitate to use the word transformational — it is. The pace of adoption will likely be far faster than prior technological transformations, like electricity or the internet.

JPMorgan Chase Technology Spending 2026: What $19.8 Billion Buys

Here's the number that actually matters. JPMorgan expects to spend roughly $19.8 billion on technology in 2026, according to reporting on the bank's JPMorgan Chase technology spending 2026 plans — a sharp increase from prior years. For context, Dimon said in October 2025 that the bank was spending about $2 billion annually on AI initiatives alone. That number has almost certainly grown.

Nineteen-point-eight billion dollars. That's not a pilot program. That's not an 'AI task force.' That's a full-scale infrastructure overhaul — cloud migration, data pipelines, model deployment — and it's happening right now, at one of the most systemically important financial institutions on the planet. Every fintech startup chipping away at JPMorgan's margins just got a reminder of the moat that capital builds.

The risks Dimon flags are real too, to his credit. Deepfakes targeting customers, AI-generated misinformation, cybersecurity vulnerabilities. He's not pretending the technology is without downside. His framing — that overreacting and underreacting are both mistakes — is sensible. The problem is that 'rigorous preparation in advance' looks different depending on whether you're the bank deploying the tools or the employee whose job disappears because of them.

Does AI 'Create Jobs' or Is That Just What They Have to Say?

Dimon said it plainly: AI 'will definitely eliminate some jobs, while it enhances others.' He promised JPMorgan will have 'definitive plans on how we can support and redeploy our affected workforce.' That's the kind of sentence that sounds reassuring in an annual letter and less reassuring when you're the analyst whose workflow just got replaced by a model.

The broader picture isn't just JPMorgan. On the same week Dimon's letter circulated, OpenAI industrial policy paper was released — titled 'Industrial Policy for the Intelligence Age' — calling on governments worldwide to rethink taxation, labor protections, and social safety nets in anticipation of AI reshaping white-collar employment at scale. OpenAI, the company selling the tools doing the displacing, is now urging policymakers to cushion the landing. Draw your own conclusions about what that signals.

Anthropic CEO Dario Amodei went further back in January, saying AI could eliminate up to half of entry-level professional jobs within five years. Software engineers at Anthropic itself, he noted, are already delegating most of their coding to models. 'We might be six to 12 months away from when the model is doing most, maybe all, of what software engineers do end-to-end,' he said at the time.

JPMorgan isn't unique here — every major bank and enterprise tech firm is running the same calculation. Deploy AI, reduce headcount over time through attrition, call it productivity gains. The 'new jobs in cybersecurity and AI' argument is true in aggregate but cold comfort when the timeline for those roles is years and the displacement is quarters. Dimon knows this. His letter is diplomatic about it, but $19.8 billion in tech spending speaks louder than any reassurance.

AI will definitely eliminate some jobs, while it enhances others. Our firm will have definitive plans on how we can support and redeploy our affected workforce.