Seems optimistic when there is very little intrinsic stickness due to learning the UI or network effects. Perhaps a little bit chat history - but not 285 billions worth.
Also completely ignoring the fact that most devices things will start to come with the same features directly built into the device/app - and the largest market will be as a commodity backend api that the eventually users won't know or care if it's a google or openai model.
As I see it, they need to be doing stuff nobody else can ( in either price or performance ), otherwise it's hard to justify the valuation.
