Source: Cui, Lim, Lu, Ng & Zhao (2026) — “CEO Empathy and Corporate Outcomes”
This is one of those risks that sounds abstract until you imagine explaining it after the fact. Then it suddenly becomes very concrete, very expensive, and very difficult to hide behind a slide deck.
CEO empathy is one of those leadership traits that everyone discusses but no one can measure. This paper tries to solve the measurement problem using AI — specifically by developing an Affective Mimicry Index AMI derived from CEOs’ spontaneous emotional alignment with interviewers during interviews. The intuition behind the measure: genuine empathy involves a form of emotional mirroring — the tendency to unconsciously align one’s emotional expression with the person you’re engaging with. By tracking whether CEOs’ emotional expressions shift in response to interviewers using AI-based video analysis, the researchers construct a behavioural proxy for the affective component of empathy that doesn’t rely on self-report or subjective assessment. The corporate outcomes findings are interesting precisely because they resist easy interpretation. CEO empathy, as measured by the AMI, correlates with certain positive outcomes — improved stakeholder management, better detection of emerging organisational problems. But it also correlates with some outcomes that suggest the risks of empathy in a CEO context: a tendency toward what the paper characterises as emotionally biased responses and potential misplaced leniency. This ambiguity is actually what makes the research valuable. The prior literature on empathy in leadership tends toward either “empathy is good” or “empathy has limits” without much precision about when and why.
My takeaway: the danger is rarely the dramatic thing in the headline. It is the quiet gap between knowing a risk exists and assigning someone to do something about it. Very unglamorous. Very important.