JPMorgan Keeps Adding Staff Even as AI Absorbs 40% of Some Team Roles in 2026

When Fast Company reported that Jamie Dimon said JPMorgan has slashed 40 percent of jobs in some departments because of AI, the headline landed like a warning shot for anyone working in banking. The reality Dimon described on the earnings call, confirmed by Fortune, The Next Web, and Entrepreneur, is more specific

and more useful than the number alone. AI has cut staffing by 30 to 40 percent in a handful of narrow, repetitive functions, most affected staff were redeployed elsewhere in the bank, and JPMorgan still plans to keep hiring. At Metaintro, we track where hiring is actually moving across the labor market, and the JPMorgan story is a near-perfect illustration of how AI reshapes teams rather than simply erasing them. Here is what Dimon actually said and what it means for your career.

What exactly did Jamie Dimon say about 40 percent job cuts?

Speaking on JPMorgan's second-quarter earnings call, Dimon said, "We have had discrete areas where we did reduce jobs by 30% or 40%." The word "discrete" is doing a lot of work in that sentence. He was not describing a bank-wide purge. He was pointing to specific, narrow functions where AI had automated enough of the routine workload that the team no longer needed the same number of people. According to Fortune, most of those people were offered jobs elsewhere inside JPMorgan rather than shown the door.

That framing matters because a 40 percent cut in a single team is very different from a 40 percent cut across a workforce that The Next Web reports numbers more than 300,000 people. Dimon was candid that AI both giveth and taketh away. As Entrepreneur quoted him, "It's created a lot of jobs in our company, and yeah, there are areas where it's reduced jobs a little bit." He returned repeatedly to the idea that technology historically creates new work even as it eliminates old tasks, a point worth holding onto if the 40 percent figure is keeping you up at night. If you want a grounded read on how automation actually lands inside companies, our guide to reading AI's real impact on jobs through seven signals walks through what to watch for.

Which jobs at JPMorgan are actually being automated?

The functions Dimon flagged are the ones you would expect AI to touch first. The Next Web reports that JPMorgan now runs nearly 1,000 AI use cases spanning fraud protection, marketing, and note-taking. Those are pattern-heavy, high-volume tasks where a model can process thousands of transactions or documents in the time it takes a person to handle a few. Fraud detection in particular is a natural fit, because it is essentially a giant classification problem, and marketing content and meeting notes are exactly the kind of drafting work that generative tools do quickly.

The bank is not experimenting at the edges either. The Next Web reports that roughly 150,000 JPMorgan employees now use the company's internal large language model every week. That is close to half the workforce touching AI as a routine part of the job, which tells you the technology has moved from pilot project to plumbing. The pattern here echoes what we have seen elsewhere, from engineering teams shrinking as AI absorbs work to the way the middle manager role is changing as AI flattens org charts. If your day is dominated by repetitive, rules-based tasks, that is precisely the profile automation targets first, and it is worth thinking hard about which parts of your role only a human can do.

It helps to notice what these functions have in common. Fraud protection, marketing copy, and meeting notes are all jobs where the output is high-volume and the rules are relatively stable, which is exactly the terrain where a model earns its keep. What is far less exposed is the work that sits on top of those tasks, the judgment calls about which fraud alerts actually warrant a client phone call, the strategy behind a marketing campaign, and the decision about what a meeting means for a deal. Dimon's own framing suggests the bank is automating the production of work while keeping and even expanding the people who interpret it. If you can move your own value toward that interpretive layer, you are moving in the same direction JPMorgan is pointing its 20 billion dollar technology budget.

Why does Dimon say AI will not slash costs the way Wall Street expects?

Here is the part of Dimon's message that got less attention than the 40 percent number but may matter more. He pushed back hard on the idea that AI will let banks dramatically cut costs and fatten their profit margins. As Fortune reported, his logic is competitive rather than technical. "You don't uniquely benefit from AI," he said, because in a competitive capitalist market every bank will deploy the same tools to do a better job for customers. Any efficiency you gain gets competed away as everyone else gains it too.

He backed the point with a memorable line. "If that were true, our margins would be 80 percent today," Dimon said, referring to the decades of computerization that were also supposed to make banks radically cheaper to run and never did on the scale predicted. He was equally blunt about the economics of AI itself. Entrepreneur reported that JPMorgan has spent about 2 billion dollars on AI and seen roughly 2 billion dollars of benefit, a one-to-one return rather than a windfall. In his words, "We have shown that for $2 billion of expense, we have about $2 billion of benefit." For workers, the implication is oddly reassuring. If AI is not a magic margin machine, the incentive to gut headcount purely to please shareholders is weaker than the doom headlines suggest.

How big is JPMorgan's AI investment?

The scale of JPMorgan's technology commitment gives the 40 percent story its context. Fortune reports that the bank spends close to 20 billion dollars a year on technology, one of the largest tech budgets of any company on earth. Entrepreneur notes that JPMorgan Chase sits on roughly 3.9 trillion dollars in total assets, so the firm has both the money and the data volume to push AI further than almost any rival.

That spending is starting to show up as a genuine cost line rather than a rounding error. The Next Web reports that CFO Jeremy Barnum described token spending, the cost of actually running the models, as trivial through the end of 2026 but flagged that he expects a meaningful acceleration in the second half of the year as usage climbs. In plain terms, the bank is still figuring out which models to run where and how much it will cost at full scale. That uncertainty is one more reason to treat Dimon's job-cut numbers as the leading edge of a still-evolving shift rather than a finished verdict. For a broader look at how the market as a whole is absorbing this spending, our breakdown of what AI skills employers actually want tracks where the money is turning into hiring.

Is JPMorgan actually shrinking, or just changing shape?

This is the question that should reframe the whole story. JPMorgan is not a company in retreat. Fortune reports the bank posted 21.2 billion dollars in net income in the second quarter, up 41 percent year over year, with every business line delivering record revenue and investment banking fees rising 30 percent to 3.3 billion dollars. A firm printing record profits is not cutting jobs to survive. It is reallocating talent toward the work that generates the most value.

Dimon was explicit about where the headcount is heading. Fortune reported that the bank will likely hire fewer traditional bankers and more AI people going forward. That is the real signal buried under the 40 percent headline. The composition of the workforce is changing, not its overall trajectory. We have already covered how JPMorgan keeps hiring aggressively where it sees opportunity, including its move to add 75 bankers for small-cap dealmaking, and the same dynamic is visible across the sector in stories like Citigroup's restructuring under Jane Fraser and Santander's Asia banking reshuffle. The banks that are cutting in one place are almost always building in another.

What does the JPMorgan shift mean for your career?

The practical lesson is that AI fluency has become a core job skill in finance, not a specialist niche. When 150,000 of your colleagues are using an internal AI model every week, per The Next Web, the person who knows how to get useful output from those tools is worth far more than the person who avoids them. That is true whether you work at a megabank or a ten-person startup. The workers most exposed to the kind of cuts Dimon described are the ones whose entire role is a repeatable process that a model can now do faster.

The defensive move is to make yourself the person who directs the AI rather than the person the AI replaces. Concretely, that means learning to use the tools already sitting on your desk, understanding what your function is actually for beyond the paperwork, and shifting your energy toward judgment, relationships, and the messy human decisions that models still handle badly. Our guide to making your role AI-resilient breaks this down step by step, and if you are wondering whether your specific job is exposed, Will AI take my job? is a candid place to start. The human skills question matters too, because as automation spreads, the abilities machines cannot copy become more valuable, a trend we unpack in human skills that get more valuable as AI spreads.

There is a hidden opportunity in the redeployment detail too. When JPMorgan cut roles by 30 or 40 percent in some teams, it moved most of those people into other parts of the bank rather than laying them off, according to Fortune. That is a template you can use even if your employer is less generous. The workers who land those internal transfers are almost always the ones who had already built transferable skills and relationships across the organization, so they were easy to place somewhere new. Staying narrowly expert in a single automatable task is the riskiest position to hold. Building a broader base, volunteering for the projects where AI is being rolled out, and making yourself known outside your immediate team are the moves that turn a restructuring from a threat into a reshuffle you survive.

How do you build AI skills fast enough to stay ahead?

You do not need a computer science degree to stay employable in an AI-heavy workplace. What employers reward is the ability to use these tools to do your existing job better, and that is learnable in weeks, not years. The fastest path is to pick the two or three tasks you do most often and figure out how AI can compress them, then keep pushing until the tool is genuinely part of your workflow. Dimon's own framing supports this. He kept insisting that technology creates jobs even as it eliminates tasks, which means the people who thrive are the ones who move up the value chain as the routine work gets automated away.

If you want a structured on-ramp, our 90-day plan to build AI skills lays out exactly what to learn and in what order, and how to upskill for AI without going back to school covers the low-cost, no-degree route. For a longer horizon, the skills that keep you hireable are the ones we track in AI skills that keep you employable in 2026. The banks are telling you exactly what they will pay for next. Dimon said JPMorgan wants more AI people. The smart response is not to fear that sentence but to become one of those people inside whatever industry you already work in.

Related Articles

• JPMorgan is hiring 75 bankers for small-cap M and A in 2026
• Citigroup restructuring under Jane Fraser and what it means for bank jobs
• Santander Asia restructuring and banking jobs in 2026
• Engineering teams are shrinking as AI absorbs the work
• The middle manager role is changing as AI flattens org charts
• Reading AI's real impact on jobs through seven signals
• Making your role AI-resilient
• Will AI take my job?
• A 90-day plan to build AI skills
• AI skills that keep you employable in 2026
• AI skills employers actually want
• Human skills that get more valuable as AI spreads

People Also Asked

Q: Did JPMorgan really cut 40 percent of its jobs because of AI?

A: No. Jamie Dimon said on the bank's second-quarter earnings call that AI reduced jobs by 30 to 40 percent in a few discrete, narrow functions, not across the whole company, and that most affected employees were offered other roles inside JPMorgan. The bank still employs more than 300,000 people and posted record revenue, so the figure describes a shift in specific teams rather than a company-wide layoff.

Q: How much does JPMorgan spend on AI and technology?

A: JPMorgan spends close to 20 billion dollars a year on technology and runs nearly 1,000 AI use cases across areas like fraud protection, marketing, and note-taking. Dimon said the bank has spent about 2 billion dollars specifically on AI and seen roughly 2 billion dollars of benefit, and around 150,000 employees now use its internal AI model every week.

Q: Will AI make banking jobs disappear?

A: Not wholesale. Dimon argued AI both creates and eliminates jobs, and said JPMorgan will hire fewer traditional bankers but more AI-focused people going forward. The likeliest outcome is that banking roles change shape rather than vanish, which makes building AI fluency inside your current job the strongest defense.

Future-proof your career before the next AI shift hits your team. Metaintro tracks where hiring is actually moving so you can build the skills employers are paying for right now, not the ones they wanted last year. Create your free Metaintro profile and get job matches and career intelligence built for an AI-driven market.