As AI companies confront rising safety concerns and growing commercial pressures, questions emerge about whether calls to slow development reflect genuine risk management, economic interests, or both.