AI Empathy in Customer Service

When things go wrong, can a machine's empathy make them right?

We investigate whether empathic AI voice assistants can repair customer interactions when technology or service fails. Our research examines how the timing and form of AI empathy influence frustration, aggression, trust, satisfaction, and service recovery.

About the programme

A desk service bell, drawn in small squares

AI voice assistants now handle customer service in high-stakes contexts like banking, yet interaction failures remain common, frustrating users and eroding trust.

We study empathy as an emotional buffer: how an empathic response at the moment of failure regulates customer emotions, protects satisfaction, and defuses aggression. The insight we pursue is precise timing: understanding when empathy repairs a broken interaction, and when it merely decorates one.

Working paperSSRN

No Hard Feelings: The Protective Power of AI Empathy During Service Interaction Failures

Read the work sample

Related explainers

DefinitionWhat is AI empathy?A machine’s ability to recognize how someone feels, understand why, and respond in a way that helps.
MethodsHow do you measure empathy in a machine?Researchers score what the system says, ask people what they perceived, and track what they do next.