The Biggest Potholes in the Path of Car-Sharing
blog.getscaffold.com
blog.getscaffold.com
I think that a better approach is more development of public transportation solutions, including the inclusion of more bike-friendly commuting features in cities. If a combination of bike commuting and public transportation became the norm for US city-dwellers, then it would be far more feasible to cut back in the number of cars (and perhaps eventually get to a point where car sharing was more culturally and logistically feasible).
(This is the perspective of someone who doesn't live in downtown NYC/SFO/LA/Seattle; I'm sure going car-less seems a bit more feasible in those very concentrated locations)
Im probably in the minority but I'm not sharing my car or my home to people, even though I know inherently most people are ok.
Even assuming everything goes splendidly with Google's self-driving cars, it is going to take a long time to move from that to fully autonomous, driverless cars. This isn't going to get disrupted until the mid-2020's, IMO.
On the other hand, using the Internet to efficiently aggregate low-use, high-cost resources is a great way to build a business. It is at the core of cloud computing, Air BnB, and (successful) Groupon promotions. I'm sure others exist as well.
http://www.zipcar.com/ (mentioned in the article)