OpenAI has temporarily halted development on its new AI model, “Astra,” due to significant security concerns. The company acknowledges that Astra possesses what it terms “critical cyber capabilities,” raising fears about potential misuse. Details regarding these capabilities remain limited, fueling speculation about the model's offensive potential. This pause affects certain aspects of Astra’s development, indicating it’s not a complete shutdown, but a strategic reassessment. OpenAI is prioritizing safety protocols and further evaluation before resuming work. The decision highlights the growing anxieties surrounding increasingly powerful AI systems and the need for robust safeguards. This move underscores the challenges of balancing innovation with responsible AI development in a rapidly evolving field.