Visible chains of thought are a safety advantage for AI, but that transparency is slipping away | AIChainDay