For decades, juniors learned the job through the unglamorous work — research, data cleanup, first drafts marked up in red. AI now does much of that. So how do they learn?
Bryan Hancock and Charlotte Seiler of McKinsey propose what they call the answer-key model. The employee attempts the task first. The AI then produces its own version. Finally, a manager walks through the gap between the two.
One real estate firm put the idea into practice. Junior staff walked neighborhoods and studied traffic patterns to build a market assessment by hand, then held it up against the AI's version. The gap became the lesson.
The model turns AI from a threat to junior learning into a training tool, and it places the weight on the manager. The comparison teaches when someone sits down and asks, "Why did you miss this?" That conversation is the apprenticeship. It is also what turns two documents into a lesson.
The idea has deep roots. Apprentices in many crafts have long learned by trying first and having a master correct the work. The answer-key model gives that old pattern a new partner.
Talent Takeaway
Pick one recurring task on your team. Next time, have your newest person do it without AI, then compare — together.
Sources
Hancock, B., & Seiler, C. (2026, July). Building expertise in the age of AI: Who trains the next generation? McKinsey Quarterly. https://www.mckinsey.com/capabilities/people-and-organization/our-insights/building-expertise-in-the-age-of-ai-who-trains-the-next-generation
