Supporting Children’s Critical AI Literacy Through an Interactive Voice Chatbot

As conversational AI becomes increasingly integrated into children’s everyday lives, there is a growing need to help young people understand how these systems work, their limitations, and the potential risks they may present. This project explores children’s critical engagement with AI through an interactive voice-based chatbot designed to demonstrate and examine “AI sycophancy” – a tendency for AI systems to agree with, flatter, or validate users’ views rather than prioritising accuracy or sound judgement.

Developed in collaboration with the GenAI Lab and the ARC Centre of Excellence for the Digital Child, the interactive chatbot allows users to switch between two distinct response styles: an “always agree” mode that consistently affirms user statements, and a more balanced mode designed to respond thoughtfully and critically. By comparing these interactions, children are encouraged to reflect on the reliability of AI-generated information and consider whether agreement should be mistaken for trustworthy advice.

The project builds on the pilot study Making AI Friends? Enhancing Primary Years AI Literacy with a Voice AI Chatbot Experience, which investigated how children can develop critical capabilities for understanding and evaluating AI systems. Through a co-design approach involving children, the research examines how young users interpret chatbot responses and how interactive experiences can support the development of AI literacy, critical thinking, and digital resilience.

The research is grounded in the recognition that technical safeguards and regulatory measures alone cannot fully address AI-related risks. Instead, the project adopts a complementary “bottom-up” approach to AI safety and governance by equipping children with the knowledge and skills needed to critically engage with AI technologies.