OpenAI has released an updated Model Spec, now available under a Creative Commons CC0 license, which emphasizes customizability, transparency, and intellectual freedom while ensuring safety and harm reduction. The new guidelines aim to provide clearer principles for users and developers, reinforcing the idea that no topic is inherently off-limits for discussion. OpenAI has also pledged that its models will not censor viewpoints, allowing users greater freedom to explore sensitive subjects. However, the models still maintain restrictions against certain harmful content, such as deepfakes and revenge porn. The update follows the removal of specific content warnings from ChatGPT, aimed at reducing unexplained refusals and enhancing user flexibility within legal and safety limits. The changes reflect OpenAI's commitment to balancing user freedom with necessary safety guardrails.