Bank of America Bets on Interns as AI Job Fears Grow

Bank of America answered artificial intelligence (AI, software that can generate, classify and assist decisions) job fears on June 3: the bank said it will welcome nearly 4,000 summer interns and full-time campus recruits, including 2,000 interns reported by Bloomberg News, while it keeps pushing automation into coding, support, research and digital banking.

The question shifts to the work itself. A college hire walking into a bank with AI assistants on employee desks is joining a training system where some starter tasks have already moved to software.

Bank of America Keeps the Campus Door Open

The official June campus hiring release said the group will be recruited from more than 500 colleges and universities to support clients and long-term growth, with the bank calling out client-facing and technology talent. Bank of America also said previously announced entry-level channels remain active, including its military veteran program and community college hiring.

Our approach to hiring is intentional and long term.

Sheri Bronstein, chief people officer at Bank of America, said in the release that the company is looking for skills, potential and a career mindset. Bloomberg News reported a precise split inside the campus class: 2,000 summer interns and 2,000 full-time college recruits across eight business lines, with Josh Bronstein, the bank’s head of global talent, saying the overall total matches last year’s campus hiring level.

That puts the hiring pledge inside a bank already using AI as an operating tool. It also puts the intern class at the center of a practical question for Wall Street: how much junior labor does a large bank still want when software can draft, summarize, search and triage?

The Automation Footprint Is Already Large

On Bank of America’s technology and AI disclosures, the bank says it spends $13.5 billion annually on technology, with more than $4 billion allocated to new initiatives including AI. It also says it has invested $118 billion in technology over the last decade and $1.5 billion in data capabilities over the last five years.

The use cases are production tools. Erica, the bank’s AI virtual financial assistant, has passed 3.4 billion client interactions since launch and is used more than 2 million times per day, according to the same page. The bank says 95% of employees have adopted Erica for Employees, while technology staff have seen efficiency gains of more than 20% through coding assistance.

For a summer analyst, those figures describe the desk before the first assignment arrives. The employee help system, the client assistant, the coding aid and the research search layer are already part of the working environment.

Campus Programs Still Feed Full-Time Roles

The training logic has survived because banks still convert internships into hires. Bank of America’s campus recruiting process says internship and full-time recruitment runs on a rolling basis once applications open, with interviews often beginning before deadlines and programs closing as positions fill.

Its recruiting material says summer internships typically last 10 weeks and attract students who aim to start full-time work the following year. The same page describes the summer internship class as the primary source of hiring into full-time campus programs.

That pathway gives the bank time to evaluate how a student works with clients, colleagues, controls and software. It also gives the student time to learn how a regulated institution handles risk before any full-time offer turns into responsibility for live accounts or internal systems.

The First Rung Has Become Harder to Reach

Outside the bank, early-career data cuts in different directions. The National Association of Colleges and Employers (NACE, a nonprofit that tracks college hiring) projected in its spring hiring update for college graduates that employers expect to raise Class of 2026 hiring by 5.6%, while demand for AI skills in entry-level jobs has nearly tripled since fall.

The Stanford Digital Economy Lab reached a colder finding in its AI labor-market study on young workers. Researchers found a 16 percent relative decline in employment for workers aged 22 to 25 in the most AI-exposed occupations after controlling for firm-level shocks.

The pressure reaches beyond Wall Street. Oton Technology has tracked executives softening public forecasts in AI chiefs walking back job apocalypse warnings, while H1-B returnees entering an AI-reshaped India job market showed experienced workers crowding into markets where entry points have already changed.

Junior Work Gets Rewritten at the Desk

Large banks have a reason to preserve junior cohorts after adopting productivity tools. Bank work carries controls, documentation, client rules and escalation paths. Someone has to learn those pathways before they can approve, challenge or correct work later in a career.

At desk level, the first months likely look different:

• Client preparation starts faster when AI summarizes account history, research and service notes, then a human checks the exceptions before a call.
• Software work moves quicker when code assistants write drafts, but bank code still goes through review, security testing and production controls.
• Operational support changes when internal assistants reset devices or surface policy answers, while managers still assign responsibility for a client outcome.
• Training rotations expose hires to credit, markets, wealth and risk work learned through supervision, feedback and mistakes caught early.

The junior seat still carries risk. When a task becomes repeatable, a bank can shrink the number of people assigned to it or combine it with work that once belonged to another role. An intern class can survive that shift, but the job after the internship gets more demanding.

Headcount Through Attrition

The headcount line already points lower. Bank of America’s first-quarter supplemental filing listed 212,134 employees at March 31, 2026, down from 213,207 at Dec. 31, 2025 and 212,732 at March 31, 2025.

The one-quarter decline was 1,073 employees; the 12-month decline was 598. Bloomberg News reported that the bank is using AI to lift productivity while managing staffing through natural attrition. Natural attrition means employees leave or retire and management decides which seats to refill.

Campus hiring and attrition can run together. A bank can bring in interns for a future analyst bench while allowing mature roles or manual support work to fade as employees leave. That is why the intern number deserves attention without being treated as a full answer to job-loss fears.

The New Entry-Level Deal

Students do get a signal from this announcement. Bank of America is still opening a formal path onto the payroll, and new hires now have to prove they can work with AI under bank controls. AI fluency in a bank means prompt use, source checking, privacy discipline and an instinct for when a generated answer needs human review.

The bank gets a staffing hedge. It keeps a cohort that can learn its controls and client base while automation removes some of the repetitive work that used to fill an analyst’s first months.

The summer class starts as a hiring answer; its conversion into permanent roles will show how much room the first rung still has.