Since ChatGPT was launched in November 2022, Generative AI (Gen AI) has captivated the public imagination. Just over a year later, millions of people are using OpenAI’s chatbot and similar tools for learning, entertainment and work. Gen AI’s significance in our economies and societies is expected to continue to grow.

Much space has been devoted to analysing and tackling the risks that Gen AI will bring. This is not without reason – for all its promises, Gen AI could make the creation of problematic content easier. We should prevent Gen AI from generating child exploitation material, detailed instructions for mass shootings, and other abhorrent or dangerous content.

This paper reviews the policies of six major chatbots – AI21 Labs Chat (AI21 Labs), Gemini (Google), ChatGPT (OpenAI), Claude (Anthropic), Coral (Cohere), and Pi (Inflection) – and examines the type of content they prohibit. It focuses on Gen AI models with web interfaces, referred to as 'chatbots'. The assumption is that most ordinary users will use Gen AI through these tools.