When the CEOs of Wall Street's largest banks begin discussing artificial intelligence, the entire financial world takes notice—from social media chatter to stock market reactions. Their predictions on how generative AI will transform work are powerful enough to sway stock prices, reshape corporate strategies, and set the tone for American business. Plans and increasingly frequent actions by banks around generative AI reveal how this technology will augment, and in some cases, replace human employees. Many major banks each employ over 100,000 people, while the broader commercial banking system protected by the FDIC employs nearly 2 million individuals. Everyone is potentially affected, from bank tellers to dealmakers earning hundreds of millions annually, and the vast teams of software engineers turning generative AI dreams into reality.

To gain deeper insight into the perspectives of banking professionals, we have compiled the most striking statements regarding workforce size in the AI era. Jamie Dimon, CEO of JPMorgan Chase, is known for his blunt style, which means he is unafraid to acknowledge that job cuts are becoming a reality. "AI is going to eliminate jobs," Dimon conceded at a conference earlier this month, "and people shouldn't keep avoiding the truth." This sentiment echoes his remarks earlier this year at an employee town hall in Columbus, Ohio, where he stated AI would "change some jobs"—whether acting as a "co-pilot," handling "repetitive tasks," or directly replacing roles. In a more recent interview, Dimon suggested that as AI adoption continues, JPMorgan's headcount could remain stable or even grow, provided "we do it right."

The core of this promise lies in boosting efficiency, which Dimon elaborated on at the 2024 Alliance Bernstein conference: "It will impact every job, every application, every database, dramatically increasing human efficiency. Think of all the time spent clicking and taking notes—soon, a button press will automatically summarize a meeting, saving wasted hours." He also explained how efficiency gains could create more jobs in cybersecurity: "We use it for risk and fraud recognition, but bad actors use it too. So we must use AI to fight them, continuously enhancing our cyber defenses." While Dimon envisions a future, decades away, where workdays and rest days might be equal due to high efficiency, he acknowledged that near-term hiring momentum will likely slow.

JPMorgan President Marianne Lake was even more direct about the structural changes to the workforce. At the Goldman Sachs Financial Services Conference in early December, she noted that productivity in operations could surge 40% to 50% over the next five years. However, she emphasized this would not trigger mass layoffs but rather slow net headcount growth—through automation, digital assistants, and self-service tools, the workload each employee can handle will increase significantly. CFO Jeremy Barnum added that the company is "asking all lines of business to hold headcount flat where possible, with a greater focus on efficiency."

The clearest statement from Goldman Sachs CEO David Solomon regarding AI's impact came from a memo he co-authored this year with President John Waldron and CFO Denis Coleman. This memo, announcing the third phase of the firm's OneGS cross-divisional plan, stated that AI would drive efficiency gains, implying a slowdown in hiring and a reduction in roles—a practice Goldman is familiar with, as it conducts annual personnel optimizations. "We are asking all lines of business to hold headcount flat where possible, with a greater focus on efficiency," the memo read, framing this as part of a broader strategy to find the "optimal team structure" and enhance "agility." When the memo was reported, a Goldman spokesperson clarified the firm still expects total headcount to be higher by the end of 2025, noting a 5% global workforce increase to about 48,000 in the third quarter. Slowing hiring and increasing headcount are not contradictory—the focus is shifting to hiring the *right* talent. "We need more high-value people," Solomon said in an October interview, "We have the capacity to hire more high-value people to expand our business footprint." He maintains that AI will ultimately lead to headcount growth over the next decade.

At a conference last month, Solomon noted, "Some areas will clearly see significant staff reductions, but I prefer to think our teams will free up more time to serve clients." He highlighted that the most immediate AI impact would be in software development. Speaking more broadly earlier this month, he stated, "Disruption is inevitable, but our economy is remarkably resilient and flexible. Looking back centuries at technological change, humans have always adapted, started new businesses, and created new jobs. This will be no different." Solomon believes AI's adoption could be faster than transformative technologies like the internet or electricity, and that professional roles might be hit hardest: "Demand for some white-collar office jobs will decrease, but other parts of the economy will absorb that workforce."

Citi CEO Jane Fraser outlined her macro thoughts on AI in banking in a LinkedIn post as early as 2023. Like her peers, she foresees a major transformation: "In the near term, generative AI will massively boost productivity; in the long run, it could fundamentally alter functions across banking and the industry—from writing code and client onboarding to fraud detection, market research, and compliance." On a recent earnings call, she provided examples of AI-driven efficiency gains: "This year, AI-powered automated code reviews have exceeded 1 million, significantly boosting development efficiency. This single initiative saves about 100,000 work hours per week." She also noted AI helps service teams resolve inquiries faster, assists wealth advisors in providing more personalized advice, and mentioned a pilot using agent-based AI for more complex tasks. When asked this month if productivity gains from AI would lead to job cuts, Fraser admitted her "worry" is that AI might "squeeze the job market before it creates returns." However, with global AI adoption rates currently around just 10%, she said widespread impact on employment is distant, suggesting a 50% adoption rate is needed to see clear trends.

Since Charlie Scharf became CEO of Wells Fargo in 2019, the bank has reduced its workforce by nearly a quarter, and he believes this trend will continue. "Looking ahead, we will likely continue to shrink in size... hopefully achieving as much as possible through natural attrition." He framed the reduction as a necessary outcome of focusing on areas with "terrible efficiency and bureaucracy." The bank's growth was constrained by a $1.95 trillion asset cap from 2018 until this June. In an interview, he specifically countered arguments that AI won't reduce jobs: "The opportunity from AI is very significant. Anyone who says AI won't lead to headcount reduction either doesn't understand what's happening or isn't being direct." In early December, he clarified further, stating most people know the reality but are "unwilling to say it, because no one wants to publicly admit future reductions are needed—it's a difficult thing to say." He revealed that generative AI tools have already boosted engineer productivity by 30%-35%. While programming jobs haven't been cut yet, the technology will ultimately allow the bank to do more with fewer people across compliance, legal, call centers, and even banking teams.

Bank of America set a new industry standard with its company-wide minimum wage of $25 per hour. Although CEO Brian Moynihan acknowledged in a September interview that applying generative AI has led to downsizing in some departments, the bank is focusing on retraining employees for roles that large language models cannot replace. "It's about redeployment and reskilling," he stated, "We have to be much more focused on multi-dimensional training than we were two or three years ago." At the Goldman Sachs conference this month, he noted the bank has kept overall headcount stable through redeployment rather than new hiring, with AI playing a central role in handling incremental work. Citing Erica, the bank's AI assistant for clients, as an example, he demonstrated tangible results: in November, the bank had 1.4 billion digital interactions with customers, which "is now estimated to be the equivalent of saving about 11,000 full-time employee roles."