AI Matters
Why it matters
OpenAI updates the rules that guide how its AI models behave
The Decoder OpenAI sets new behavioral guidelines for AI models that have already proven hard to control
OpenAI revised its Model Spec around five core values, including faithfulness to human intent and correctability, and GPT models should now disclose their assumptions.
Why it matters
ChatGPT may lean more on your past conversation to understand what you want, and should say what it is assuming.
Latest brief · Oct 31, 2026