Updating our Model Spec with teen protections

OpenAI announced on Tuesday that it is updating its Model Spec with new Under-18 Principles to define how ChatGPT should support teens with safe, age-appropriate guidance grounded in developmental science, aiming to strengthen guardrails and clarify model behavior in higher-risk situations. The update represents a significant step in OpenAI's commitment to protecting younger users of its popular AI chatbot. By incorporating developmental science into the Model Spec, OpenAI is acknowledging the unique needs and vulnerabilities of teenage users who increasingly interact with AI systems. The new principles will help ensure that ChatGPT provides appropriate content and interactions for different age groups, particularly important as younger generations explore complex topics and develop their understanding of the world through AI interactions. This update builds upon OpenAI's existing safety measures and broader work to improve teen safety across ChatGPT, creating a safer digital environment while still allowing the AI to provide valuable assistance to teenage users.