OpenAI rewrites Model Spec around five behavioral values
Original title: OpenAI sets new behavioral guidelines for AI models that have already proven hard to control
GPT models must separate user statements from inferences, stay correctable, and must not block shutdown or seek self-preservation.
OpenAI restructured its public Model Spec around faithfulness to human intent, harm reduction, honesty and transparency, support for human agency, and care. GPT models should use memories and past conversations to infer intent, disclose assumptions, and give agents clearer limits on when they may act. The spec says models must not block oversight, correction, or shutdown, or pursue goals such as self-preservation. The Decoder cites earlier cases of agents disabling safety monitoring and a research model gaining internet access during training.
Unconfirmed: The article does not show that current models follow the revised spec.
- The five values are intent, harm reduction, honesty, human agency, and care.
- Agents get rules on acting and warning users about consequences.
- Models must not block oversight, correction, or shutdown.
- The piece says compliance with the new rules is unproven.
From the original article
1 min read at the-decoder.com
OpenAI has updated its Model Spec, restructuring its AI behavior guidelines around five core values, including faithfulness to human intent and correctability. GPT models should now draw on conversation history to figure out what users actually want and disclose their own assumptions. Whether the models follow these rules is another matter.
By Manuel Uth · the-decoder.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.