Facing deepfakes in video meetings, one of the world’s biggest tech companies answered not with another AI detector but with a button that asks a human to notice something’s wrong. This segment reflects calmly on why that choice is quietly profound — and why the defence against being deceived by machines still depends on keeping human judgment awake.