How AI Is Reshaping Post-MBA Careers: A Role-by-Role Exposure Map
Updated: 3 days ago
There are two stories about AI and careers running at the same time, and most advice only tells you one of them. The first is the one everyone has heard: AI automates tasks, junior work disappears, be afraid. The second gets less airtime: PwC's 2026 AI Jobs Barometer found that the roles most reshaped by AI are growing twice as fast as the ones AI barely touches, and paying 42% more in wage growth on top of that. Both stories are true at once, because they are describing different halves of the same job, and the actual skill an MBA needs right now is telling which half of a target role you are looking at.
That distinction, task-level exposure versus job-level disappearance, is where most career advice on this topic goes wrong. Below is a function-by-function look at where the exposure actually sits across the roles MBA graduates take, drawn from the research that has tried to measure this directly, and what that means for how you should be reading a job description, an internship, or an elective this year.
Two Kinds of AI Exposure, and Why the Difference Matters
Goldman Sachs's widely cited 2023 analysis measured something specific: the share of tasks within an occupation that generative AI could plausibly perform. Office and administrative support came out highest at 46%, legal work at 44%, and business and financial operations, the category closest to most MBA-track jobs, at 35%. That is a task measure, not a headcount forecast.
PwC's newer framing adds the missing half. It splits roles into "professionalised" jobs, ones reshaped to demand more human expertise even as AI takes over pieces of them, and "democratised" jobs, ones AI makes easy enough for non-experts to do. The professionalised roles are the ones growing and paying more; the democratised ones are growing slower and paying less, and the difference has nothing to do with how much AI touches the role and everything to do with which piece of it AI touches. A role can score high on Goldman's exposure measure and still be a great bet, if the exposed piece is the low-value piece.
We made the case for why the MBA itself still holds up under all of this in Relevance of MBA in Age of AI. This piece goes one level deeper, into which specific post-MBA functions carry which kind of exposure.
Where the Exposure Actually Sits, Function by Function
The table below maps nine common post-MBA destinations to an estimated range of task-level AI exposure, synthesized from Goldman Sachs's 2023 occupational analysis, the World Economic Forum's Future of Jobs Report 2025, and PwC's 2026 AI Jobs Barometer. Treat every number as directional. These are estimates of how much of a role's current task mix AI can plausibly touch, not a probability that the role itself gets eliminated, and the gap between those two things is exactly the point of this whole article.
Role | What They Do | Est. AI Task Exposure* |
Data / Business Analyst | Pull, clean and interpret data to answer a specific business question for a decision-maker. | 40-50% (highest) |
Investment Banking / Equity Research Analyst | Build financial models, comps and pitch books used to advise on deals and capital raises. | 40-50% (high) |
Management Consultant | Diagnose a client problem, structure hypotheses, build the analysis, and present a recommendation. | 35-45% (high) |
Marketing / Brand Manager | Plan campaigns, write and test creative, and manage a brand's positioning across channels. | 35-40% (high) |
Operations / Supply Chain Manager | Plan forecasting, inventory and logistics decisions that keep a supply chain running. | 30-35% (moderate) |
Product Manager | Translate customer and business needs into a roadmap and align engineering, design and sales. | 30-35% (moderate) |
HR / People Operations Manager | Manage hiring, performance processes and the policies that shape how people are managed. | 25-30% (moderate) |
General Manager / Business Unit Leader | Own the P&L for a business or region and make resourcing trade-offs across functions. | 20-25% (low) |
Sales / Business Development Manager | Build and close relationships with new or existing accounts to hit a revenue number. | 15-20% (lowest) |
A few things stand out immediately. Pure data and analysis work sits at the top, because it is closest to the raw data-processing tasks every study flags as most automatable. Sales and general management sit at the bottom, for the same reason across every study: relationship-building and judgment under genuine ambiguity are the hardest things to hand to a model. Everything else, consulting, product, marketing, HR, operations, falls in a wide middle band where the honest answer is "it depends which part of the job you mean."
If consulting is on your list, we went deeper on what AI is actually doing to that specific function in Is AI quietly killing consulting?, and if product management is your target, The Rise of AI Product Management covers how the role itself is splitting into AI-fluent and AI-adjacent tracks.
Why Junior Analytical Work Is the Most Exposed, and What That Means for You
Here is the uncomfortable part for anyone entering a two-year gap or a one-year MBA specifically to break into one of these functions. The tasks AI automates first, first-draft models, first-draft decks, first-pass screening, first-cut market research, are disproportionately the tasks firms used to hire junior analysts and associates to do. That is not a hypothetical. PwC's own data shows the most AI-exposed junior roles are now seven times more likely than the least-exposed junior roles to require traditionally senior skills like leadership, which is a polite way of saying the entry-level rung of the ladder is being asked to arrive pre-promoted.
If your post-MBA plan depends on an entry-level or associate-level seat in one of the high-exposure rows above, the real question to ask isn't whether AI will take the job, it's whether the firm still needs as many people at that rung to do what's left once AI has taken the first draft. That is a headcount question, not an existential one, and it is answerable if you ask it directly in an interview rather than assuming either extreme.
If you are still mapping out which functions to target before you commit to an application cycle, Launchpad's Placement Prep Academy has a free opening module for each track, consulting, product, finance and more, which is a low-commitment way to see how placement prep for a specific function is actually changing right now.
The Skills That Move You to the Safer Side of Every Row
Across every row in that table, the tasks that survive share a pattern: they require judgment under incomplete information, they involve convincing or aligning other humans, or they involve catching what the model got wrong. None of those are new MBA skills, but they are the ones worth deliberately over-investing in relative to the technical or analytical skills that used to be the safer bet.
The most underrated of these right now is the ability to verify and push back on AI output rather than simply accept it, because the people getting the most value out of AI tools inside firms right now are not the ones who prompt best, they are the ones who can tell when the output is confidently wrong. That is a judgment skill wearing a technical costume, and it is exactly the kind of thing a case classroom or a client project is built to train, if you treat it as the point of the exercise rather than as a box to check.
What This Means When You're Choosing Electives, a Concentration, or an Offer
Don't use "AI-proof" as your filter when you're picking a concentration or an offer. Almost nothing on this list is fully AI-proof, and the roles that look safest today, general management and sales, are safest because they sit closest to judgment and relationships, not because they are somehow outside AI's reach. Pick the function based on where you want to sit relative to the judgment layer of the work, then ask specific questions in every interview about how that function's junior-to-senior ratio has actually moved in the last two to three years, since that ratio, not a headline about job losses, is the real early signal of whether a function is hollowing out from the bottom.
Quick Answers
A few things candidates ask us often:
Does a high AI-exposure percentage mean a role is disappearing?
No. It measures how much of the role's current task mix AI can plausibly perform, not the probability the role itself gets eliminated. PwC's research actually shows many high-exposure roles growing faster and paying more, because the role gets reshaped around the parts AI can't do rather than eliminated outright.
Which post-MBA function is safest from AI right now?
General management and sales roles consistently score lowest on task-exposure across Goldman Sachs, WEF and PwC's research, because they lean hardest on judgment under ambiguity and relationship-building. Safest doesn't mean untouched, though; it means the exposed slice of the job is smaller.
Should I avoid consulting or investment banking because of this?
Not on this data alone. Both sit in the moderate-to-high exposure band because their junior tasks, modeling, first-draft decks, are highly automatable, but the senior, client-facing layer of both functions is holding up. The practical move is to ask how the analyst-to-associate ratio has shifted recently, not to avoid the function outright.
Comments