Problem
Research questions and friction points this paper is trying to address.
AI-enabled toys
children as designers
large language models
generative AI
child-AI interaction
If My Toy Could Talk: How Young Children Imagine, Design, and Test AI-Enabled Toys
🤖 AI Summary
This study investigates how children design AI personas for toys and navigate the behavioral biases of generative AI. To this end, we developed ToyTalk, a code-free technology probe that empowers children aged 7–9 to assume designer roles, enabling them to independently configure and evaluate large language model (LLM)-driven toys while analyzing their design preferences and interaction strategies. Our findings reveal children’s distinct expectations regarding AI coherence, memory retention, and social roles. Notably, children predominantly designed positively oriented social personas and, when encountering output biases, largely employed corrective and retry strategies rather than engaging in system-level reconfiguration. This work provides a novel paradigm and empirical evidence for child-participatory AI design.
📝 Abstract
To investigate the design space where children might design AI chatbots for their own toys, we developed ToyTalk, a technology probe that positions children as designers of LLM-enabled toys. Children begin with a familiar toy, configure its AI-enabled version through a no-code interface, and then interact with and test the character. We deployed ToyTalk with 76 children aged 7-9 across five elementary schools in the southeastern U.S. We examine how children define their toy, probe what it becomes, and respond when behavior diverges from expectations. Children predominantly designed toys with socially positive personalities, supportive roles, and interpersonal rules. In conversation, they most often probed identity and knowledge, while also testing capabilities, memory, and relationships. When mismatches arose, children typically responded through correction, persistence, and retesting, while few returned to reconfigure the system. We discuss implications for children's design agency, testing practices, and expectations of coherence in child-facing generative AI.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Large Language Models
No-code Interface
Technology Probe
Child-Computer Interaction
Generative AI
🔎 Similar Papers
No similar papers found.
