Original title: AI Empathy as a Human Disposition in Generative AI Interaction: Development and Preliminary Psychometric Evaluation of a Cognitive–Affective–Motivational Scale
SukJae Joshua Kang, Juyoung Lee, Seong-Joo Kang · September 2026
People use generative artificial intelligence for reasoning and solving problems, but they react differently to how it acts and where it fails. Researchers wanted to measure these human reactions. They created a survey called the AI Empathy Scale for Generative AI Interaction, made up of 42 items. The survey treats human empathy toward artificial intelligence as a mix of three parts: thoughts, feelings, and motivations. It also splits these parts into self-focused and AI-focused responses. To test the survey, students answered questions before and after doing activities with generative AI. The researchers found that most empathy scores went up after these activities, except for personal distress, which stayed about equivalent. Gains were larger for understanding AI perspectives, imagining yourself in its place, intending to help, and reflecting on the interaction than for empathic concern. These results show that empathy toward AI can be measured as a personal trait.
You are teaching a friend to use a new tool. Sometimes they get frustrated, sometimes they try to understand why it failed, sometimes they want to help it improve. But you have no way to measure these different reactions consistently. Here, the researchers designed a questionnaire that captures exactly those kinds of responses when people interact with generative AI.
As generative AI becomes part of everyday problem-solving, how people interpret and respond to its limitations matters. This scale gives researchers and designers a way to measure those individual differences, which could lead to better AI systems and more effective human-AI collaboration.
The scale needs further validation with larger and more diverse groups, behavioral measures, and different contexts. One factor mostly captured negatively worded items, and the Prosocial Intention subfactor did not emerge clearly. The increase in scores after AI activities may partly reflect the activities themselves rather than a stable disposition.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
AI Empathy as a Human Disposition in Generative AI Interaction: Development and Preliminary Psychometric Evaluation of a Cognitive–Affective–Motivational Scale — SukJae Joshua Kang, Juyoung Lee, Seong-Joo Kang — Behavioral Sciences — doi:10.3390/bs16101717
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper.