The obvious one is bringing costs down by automating labour, with self-driving cars etc. There also may be a 'change the world' sort of outcome, where consumer

The obvious one is bringing costs down by automating labour, with self-driving cars etc. There also may be a 'change the world' sort of outcome, where consumers en masse like the product so much that they change their behaviour and lifestyles, even if it eventually turns out that they spend more overall than they might have before - if people stop owning cars, cease cooking at home, taking away revenue from supermarkets and auto manufacturers. Although this doesn't appear to hold up in the first analysis, second-order effects might start to appear - e.g. if consumers change their lifestyles, can you capture the value that they would have invested in owning a kitchen, or a garage and driveway? This is all very speculative, I'm not claiming Uber etc can achieve this, just that different equilibria are possible - for instance, I gather that in some Chinese cities street food is incredibly good and cheap while density is high enough to make owning good cooking facilities a real pain.