Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

"Not Human, Funnier": How Machine Identity Shapes Humor Perception in Online AI Stand-up Comedy

Xuehan Huang, Canwen Wang, Yifei Hao, Daijin Yang, Ray LC

TL;DR

This work investigates whether an AI’s machine identity can shape humor perception by designing a machine-identity-based stand-up agent and evaluating it against a baseline. Through a formative study with expert comedians, video analyses, and a controlled user study (N=32), the authors show that machine-identity humor enhances perceived humor, warmth, and presence, while also reframing human–AI interactions as computational authenticity. The findings highlight practical guidelines for deploying identity-based AI humor in education, customer service, and content creation, and propose a scalable design framework that foregrounds machine traits rather than human imitation. Overall, the work advances a principled approach to AI entertainment that leverages AI’s inherent non-human identity to produce novel, engaging, and ethically bounded performances with real-world applicability.

Abstract

Chatbots are increasingly applied to domains previously reserved for human actors. One such domain is comedy, whereby both the general public working with ChatGPT and research-based LLM-systems have tried their hands on making humor. In formative interviews with professional comedians and video analyses of stand-up comedy in humans, we found that human performers often use their ethnic, gender, community, and demographic-based identity to enable joke-making. This suggests whether the identity of AI itself can empower AI humor generation for human audiences. We designed a machine-identity-based agent that uses its own status as AI to tell jokes in online performance format. Studies with human audiences (N=32) showed that machine-identity-based agents were seen as funnier than baseline-GPT agent. This work suggests the design of human-AI integrated systems that explicitly utilize AI as its own unique identity apart from humans.

"Not Human, Funnier": How Machine Identity Shapes Humor Perception in Online AI Stand-up Comedy

TL;DR

This work investigates whether an AI’s machine identity can shape humor perception by designing a machine-identity-based stand-up agent and evaluating it against a baseline. Through a formative study with expert comedians, video analyses, and a controlled user study (N=32), the authors show that machine-identity humor enhances perceived humor, warmth, and presence, while also reframing human–AI interactions as computational authenticity. The findings highlight practical guidelines for deploying identity-based AI humor in education, customer service, and content creation, and propose a scalable design framework that foregrounds machine traits rather than human imitation. Overall, the work advances a principled approach to AI entertainment that leverages AI’s inherent non-human identity to produce novel, engaging, and ethically bounded performances with real-world applicability.

Abstract

Chatbots are increasingly applied to domains previously reserved for human actors. One such domain is comedy, whereby both the general public working with ChatGPT and research-based LLM-systems have tried their hands on making humor. In formative interviews with professional comedians and video analyses of stand-up comedy in humans, we found that human performers often use their ethnic, gender, community, and demographic-based identity to enable joke-making. This suggests whether the identity of AI itself can empower AI humor generation for human audiences. We designed a machine-identity-based agent that uses its own status as AI to tell jokes in online performance format. Studies with human audiences (N=32) showed that machine-identity-based agents were seen as funnier than baseline-GPT agent. This work suggests the design of human-AI integrated systems that explicitly utilize AI as its own unique identity apart from humans.
Paper Structure (79 sections, 1 equation, 4 figures, 4 tables)

This paper contains 79 sections, 1 equation, 4 figures, 4 tables.

Table of Contents

  1. INTRODUCTION
  2. BACKGROUND
  3. Humor and Stand-up Comedy
  4. AI for Humor and Human Perception of AI
  5. Identity as Humor
  6. Machine Identity
  7. FORMATIVE STUDY
  8. Expert Interview
  9. Participants and Recruitment
  10. Interview Protocol
  11. Thematic Analysis
  12. Video Coding
  13. Humor Strategies
  14. Performative Strategies
  15. Audience Interaction
  16. ...and 64 more sections

Figures (4)

  • Figure 1: A study screenshot showing participants and the user interface design of the AI stand-up comedy system.
  • Figure 2: The prompt and interaction design of the AI stand-up comedy system.
  • Figure 3: Study procedure including a pre-study introduction, two performance sessions, two questionnaires, and a semi-structured focus group interview. The AI comedian’s identity (human or machine) was randomly assigned and counterbalanced across performances. Participants completed one questionnaire after each performance.
  • Figure 4: Questionnaire ratings with pairwise Benjamini–Hochberg–corrected significance indicated by asterisks: (a) Perceived Humor, (b) Perceived Personality, (c) Perceived Ability, and (d) User Perception.