Almost every fresh graduate entering the labor market today shares the same anxious thoughts:entry jobs are disappearing, asAIsiphons them off. And they do have a reason to worry: U.S. entry-level job postings have fallen about35 percentsince January 2023, according to labor research firm Revelio Labs. s’
Financefeels it too. In a Stanford Graduate School of Business survey,37 percent of accountantsreported anxiety about AI’s effect on job stability. Brookings Institution research found AI could automate more thanhalf the tasks in entry-level positions. That’s roughly five times the exposure more senior roles face.
It’s true that junior work is the most automatable. Data extraction and ingestion, reconciliation work — all these operations that take weeks of manual work can be automated with high precision in a fraction of the time.
But the same shift can finally free junior specialists from repeatable tasks and allow them to focus on analysis, judgment and client exposure. That’s far better training than transaction matching, and it lets them deliver real value much sooner.
So how might automation actually impact the future of the profession? It turns out automating the mechanical floor actually gives juniors a faster route to developing the skills that matter.
How Is AI Changing Entry-Level Accounting Roles?
AI is automating repetitive tasks like data extraction and transaction matching, causing a 35 percent drop in U.S. entry-level job postings since January 2023. While this reduces routine junior hiring, it shifts entry-level work toward high-value analysis, reviewing model outputs and client exposure from day one. To adapt, firms must redesign onboarding to deliberately teach critical judgment and interrogating AI results.
What’s Threatening Junior Roles?
Traditionally, the first two years of an accounting career are built almost entirely on work that involves no analysis, but is meant to educate juniors through repetitive mechanical activities.
Pulling the trial balance, general ledger detail, bank statements and invoices out of the data room; normalizing the chart of accounts, tying the trial balance to the financials, matching invoices to revenue entries — all these tasks, which take weeks to perform manually, can be perfectly automated with high precision.
AI extracts invoice fields like vendor details, line items, and payment terms with over99 percent accuracy. If anAI modelcan match transactions in seconds, what is a first-year analyst for? This is the major fear hanging over everyone starting out in consulting and financial due diligence, and the labor market responds with a decrease in junior hiring.
Companies can no longer afford paying for juniors, who need years to get to actual work. At the same time, halting junior hiring brings a succession issue, which threatens the profession no less.
That puts urgent pressure on HR departments. They need a new framework for moving zero-experience hires, fluent inAI tools, into the profession. And first of all, they need to answer the question: How can someone start building judgment if they never had to grind through a general ledger by hand?
Rebuilding the Career Pipeline
One approach is to rebuild the junior’s path through the firm around active participation in client conversations from day one, with junior analysts fully equipped with automation tools.
As AI becomes the industry default, entry-level roles can center on reviewing model outputs and learning to spot what the model missed. That adds an analytical layer they’re capable of handling and have no reason to postpone.
There’s also a second reason to pull juniors into analytical work early. The same Stanford research found junior staff tend to accept AI outputs at face value, even when those outputs are flagged as uncertain. They also see smaller performance gains than seniors because they lack the context to challenge the machine.
A pipeline that hands juniors automated outputs without teaching them to interrogate those outputs produces accountants who can operate the tools but verify nothing. To better picture the consequences of that approach, imagine a junior on a quality-of-earnings engagement. The AI tool flagged a cluster of large invoices booked in the final days of the fiscal year as “low confidence, possible cut-off issue.”
If the analyst checks that the dates and customer names are valid, clears them, and the numbers reconcile, he misses some very important questions: Were those goods actually shipped before year-end, or did the seller pull January’s revenue into Q4 to inflate earnings?
$2 million pulled forward, on a 10x deal that wasn’t investigated properly, adds $20 million to the price. In a profession where a single mistake can cost a client seven figures, that’s a gap HR can’t ignore.
The urgency compounds because the old apprenticeship model is breaking on its own. The United States had653,000licensed CPAs as of mid-2025, and the number of people sitting for the CPA exam has fallen more than 30 percent since 2016. With three-quarters of today’s public accounting CPAs set to reach retirement age within 15 years, building proper succession and drawing young people back into the profession is now a priority.
Where Professions Are Heading
I spent more than a decade in audit and financial due diligence at the Big Four and McKinsey, and even jobs in these companies have always faced the problem of tons of tasks that needed no expert judgment. Even partners with 20 years of pattern recognition were effectively doing data entry.
Automation can finally help change the workflows once industrialized in the 1990s. By the 2030s, every level of the profession, junior to partner, will work alongside AI, and that’s why it’s inevitable.
The output isn’t only speed, but also a clear picture of where the real risk and value sit, which give more time and insight for human experts to dive deeper and create efficient strategies for businesses. Accountants can spend more time analyzing why the numbers behave the way they do, pressure-testing the story behind a target’s earnings, and shaping the strategic call on whether and how to do the deal.
It also returns time to interview management, challenge assumptions and advise clients on risk and structure — something AI is not really capable of.
Of course,firms billing by the hour will resist, but that model is a dead end. The market already knows it; pricing is moving toward outcomes, and automation is what makes outcome-based work possible. It pushes the whole profession to be measured by results rather than hours logged.
Reshape Entry-Level Roles Now
The new version of the junior accountant role is going to be built on judgment, context and client exposure. That’s a harder job — and a better one, in which skills carry weight. 60 percent of Gen Z accounting students already expect to use AI on their first day of work. They arrive fluent in the tools, so fluency alone won’t be the edge.
What’s changing is the entire shape of early-career development. Firms now have to teach judgment deliberately, from week one, because the work that used to build it has been automated away. Hiring is shifting too:91 percentof accounting professionals say graduates prefer firms that actively use advanced technology, so firms that cling to manual processes will lose talent well before they lose clients.
To survive, firms have to make structural changes: redesign onboarding around exposure, build explicit training in interrogating AI output, and move pricing toward outcomes now, while the shift is still a choice and not a forced reaction to a competitor who already made it.
