If you've ever been frustrated by a vague, generic, or suspiciously agreeable response from an AI, you might be making the same mistake a rookie interviewer makes — jumping straight to the question without setting the stage. It turns out that the skills social scientists have spent decades honing — avoiding leading questions, establishing context, triangulating sources, demanding counterfactuals — are exactly the skills that separate a mediocre AI user from a great one.
In this short guide, Slavica Stevanovic makes a compelling case: LLMs confabulate the way humans do, they're trained to please the way interviewees sometimes are, and they respond to framing just as powerfully as any research subject. The good news? You don't need a computer science degree to get better outputs. You need the same epistemic humility and methodological care that good researchers bring to every conversation. Read on to explore the parallels — and pick up a practical framework for prompting AI like a seasoned interviewer.
View Slide Deck