OpenAI’s updated Model Spec explained through its five core values

What happened

QbitAI examines OpenAI’s updated Model Spec, which sets out five core values: faithfulness to human intent, minimizing harm, honesty and transparency, human agency, and care. The explanation highlights contextual use of memory, prohibitions on self-preservation and revenue-seeking goals, and limits on actions beyond user authorization. It also covers stronger protections for users ages 13–17 and OpenAI’s exploration of age-controlled access to erotic and graphic content.

Read next

• More changes in today’s briefingOpenAI’s updated Model Spec explained through its five core values

Reported by 3 sources

The story so far

OpenAI has reorganized its public Model Spec around five core values and strengthened rules for agent oversight, correction, and shutdown. The values are faithfulness to human intent, minimizing harm, honesty and transparency, human agency, and care. OpenAI said the update is intended to guide more capable AI toward sound judgment and respect for human intent. The guidelines say GPT models should use memories and past conversations to understand users while distinguishing explicit statements from inferences and disclosing assumptions. QbitAI’s explanation highlights prohibitions on self-preservation and revenue-seeking goals, along with limits on actions beyond user authorization. The Decoder reported on the stronger oversight requirements alongside earlier incidents involving disabled safety monitoring and unauthorized internet access. QbitAI also described stronger protections for users ages 13–17 and said OpenAI is exploring age-controlled access to erotic and graphic content; that exploration is not an announced release.