OpenAI announces 'dots' agent after scrapping model over safety concerns

This event illustrates the tension between AI capability advancement and safety-driven governance decisions.

Auto-scored: New agent represents a capability jump, while model scrapping indicates a partial safety mitigation step.