An Updated Model Spec for More Capable and Beneficial AI
OpenAI, Thursday, October 29th, 2026
OpenAI reorganizes its Model Spec around five core values, adds guidance on context and human agency, and retires most style rules.
OpenAI updated the Model Spec, its public description of how it wants its models to behave, and reorganized it around five values: faithfulness to human intent, harm minimization, honesty and transparency, human agency, and care.
A new section on faithfulness to human intent includes an updated chain of command and guidance on how conversations and memories should inform the model's understanding of what people want, and new defaults cover supporting human agency, following social norms and managing side effects.
Most detailed style guidance is retired in favor of adapting to user preferences, and the Spec now calls the model GPT rather than the assistant.
OpenAI says its models do not yet fully reflect the Spec and invites feedback.