From hard refusals to safe-completions: toward output-centric safety training

OpenAI has introduced a new safe-completions approach in GPT-5, which improves both safety and helpfulness in AI responses—moving beyond hard refusals to nuanced, output-centric safety training for handling dual-use prompts.

Summary written by Kernelia from the original article by OpenAI News. The story and its rights belong to its author.