resources
What Happens to the Junior Analyst When AI Writes the First Draft?
02 Oct 2026
For decades, the first years of an analyst’s career followed a familiar script: build the model, draft the memo, format the slides, then watch a senior colleague take it apart line by line. The work was repetitive, but it was also how judgment was formed. Generative AI can now produce that first draft in minutes, and firms in consulting, banking and corporate finance are asking a quiet but important question: what is the entry-level role actually for?
The question is not only one for employers. Universities are rethinking it as well, and looking at how AI is being taught in business schools reveals a clear shift in emphasis. Less classroom time goes into producing analysis from a blank page, and more into evaluating, challenging and correcting what a model has generated. That change offers a useful clue about where the junior analyst’s value is heading.
The Apprenticeship Problem
Much of what a junior analyst learned was never written. Spotting an inconsistent assumption in a revenue forecast, sensing that a market-sizing figure looks too neat, knowing which slide a partner will question first: these instincts came from doing the groundwork hundreds of times. When a tool handles that groundwork, the output arrives faster, but the learning that came with the struggle can disappear.
This creates a paradox. Organisations still need people who will one day become experienced reviewers, yet the tasks that used to train those reviewers are shrinking. Without deliberate adjustments, a firm can end up with efficient juniors who have never had to understand why a number is right.
From Author to Editor
The most visible change is that the analyst’s starting point moves from creation to critique. Instead of writing the first version, the junior now receives one and must decide what to keep, what to rebuild and what to discard. That sounds easier, but it often demands more knowledge, not less. Editing a confident, well-formatted draft requires enough understanding of the subject to notice what is missing or subtly wrong.
AI drafts tend to be fluent and plausible. Their weaknesses are rarely obvious errors; they are outdated figures, unsupported claims, generic reasoning or conclusions that ignore a client’s specific situation. Catching these calls for curiosity and a habit of verification that cannot be outsourced.
Skills That Gain Weight
Several abilities become more valuable in this setting:
- Source checking: tracing every figure back to reliable data rather than trusting a polished sentence.
- Framing the question: a vague request produces a vague draft, so defining the problem precisely matters more than ever.
- Contextual judgment: knowing the client, the sector and the unwritten constraints that a general model cannot see.
- Clear communication: explaining why a conclusion holds, especially when it contradicts the machine’s version.
Technical skill does not vanish. An analyst who cannot build a basic model will struggle to recognise when an automated one is flawed.
What Employers Can Do
Firms that want strong future leaders can protect some of the old apprenticeship on purpose. Some ask juniors to complete certain analyses manually before comparing them with an AI version. Others require notes explaining each change made to a generated draft, turning review into learning. Senior feedback on the reasoning, not just the deliverable, also helps close the gap left by automation.
Transparency matters too. Clear internal rules on when AI can be used, how outputs must be checked and who remains accountable give junior staff a framework rather than leaving them to guess.
A Role Redefined, Not Removed
The junior analyst is unlikely to disappear, but the job is changing shape. The value lies less in producing the first draft and more in knowing whether it deserves to become the final one. Those who build that discernment early, and the organisations that give them room to develop it, will be best placed as the technology continues to evolve.
