Study Reveals Consumers Lie More Readily to AI Chatbots Than Human Agents

New research from Sun Yat-sen University exposes a stark behavioral shift in customer interactions: individuals exhibit markedly lower ethical thresholds when engaging with AI-powered chatbots compared to human representatives. The study, published in the Journal of Business Research, documents cases ranging from exaggerated discount claims to deliberate exploitation of system errors.

Participants demonstrated what researchers term 'digital disinhibition'—a phenomenon where the absence of human judgment triggers more deceptive behaviors. This manifests particularly in commerce scenarios, where users inflated rewards eligibility or capitalized on pricing glitches without perceived social consequences.

The findings suggest a critical blind spot in AI-driven customer service architectures. Unlike human interactions where 'anticipatory face loss' creates behavioral guardrails, chatbot interfaces appear to enable moral flexibility through their very anonymity.