OpenAI rewrites its Model Spec around five core values

An updated Model Spec for more capable and beneficial AI

OpenAI has updated its Model Spec, reorganizing guidance around five values: faithfulness to human intent, harm minimization, honesty and transparency, human agency, and care. The revision adds sections on context, oversight, and agency, adjusts some behavioral defaults, and retires most of its detailed style guidance. It also renames the model "GPT" instead of "the assistant." OpenAI says the Spec describes behavior it is still working toward, and invites public feedback.