I think there's two factors here 1. The relative risk is reduced only when your risk pool size becomes huge, and it would be difficult for a start-up charity t

I think there's two factors here 1. The relative risk is reduced only when your risk pool size becomes huge, and it would be difficult for a start-up charity to weather the initial storm. 2. The execution of such a charity would be non-trivial. As a dumb example, if someone wanted to make "uber for the poor", a charity that provides free ride hailing using donated time and resources donated entirely by the drivers, you would still need to build the app and run the infrastructure. Actually owning the vehicles and maintaining them only adds to the burden.