OpenAI says human intent at center of updated AI model guidelines

What happened: OpenAI updated its Model Spec, the public roadmap for how its AI models should behave, reorganizing it around five core values, with faithfulness to human intent as the foundation. Key numbers: Update announced Oct 29; framework built on five core values. Why it matters: OpenAI said current models do not yet fully reflect the updated spec and it is exploring new training methods. The move echoes rival Anthropic's Constitutional AI approach and follows scrutiny of AI alignment.

OpenAI updated its Model Spec, the public roadmap for how its AI models should behave, reorganizing it around five core values, with faithfulness to human intent as the foundation.

OpenAI said current models do not yet fully reflect the updated spec and it is exploring new training methods. The move echoes rival Anthropic's Constitutional AI approach and follows scrutiny of AI alignment.