As someone deeply involved in AI development, I believe it is crucial to prioritize safety and ethical considerations. If AI Safety Institutes identify signs of losing control, pausing training is a responsible measure. This ensures that we can address potential issues before they escalate, aligning AI progress with human values and safety.