Everything you can imagine is real. RSS

An eclectic collection of things I'm reading, looking at or thinking about.

A stream of consciousness companion to the Park Paradigm.

Archive

Jan
2nd
Mon
permalink

Unlike other car-sharing services, HiGear specialized in “high-end” auto brands only, like Mercedes, BMW, Audi, Aston Martin, Lamborghini and Tesla. It provided comprehensive liability and collision insurance and performed member screening, which included driving record checks and credit checks. It also collected a security deposit in order to “encourage safe and fair use of members’ cars.” In additional to the rental fee ($125-600, depending on the car), drivers would pay $20-40 per day in rental insurance.

These protections proved to be insufficient, however, when HiGear was targeted by a criminal ring. The group stole four cars totaling $400,000. The criminals used stolen identities to bypass HiGear’s security checks, and stolen credit cards to pay HiGear’s fees. Police have since recovered some of the cars and insurance is now processing claims for the rest, but the incident has forced HiGear to realize that this sort of thing may not be preventable in the future.

Luxury Car-Sharing Service HiGear Shuts Down Due To Theft | TechCrunch

Just wondering aloud as to whether or not they (or other such p2p marketplaces) could use a combination of social network data and algorithms to enhance their traditional KYC / vetting.  I guess it would depend on how people use these various tools, but for example I would generally - all other things being equal - be happier to rent a high value car (or other good) to people I know and follow on Facebook, LinkedIn and Twitter (with some caveats) and possibly 2nd degree connections as well (perhaps with the intermediate person acting as a reference.)  I also think things like a person’s PeerIndex score could be a relevant (information ratio enhancing) data point.

I’m not suggesting that any of these elements on it’s own would be a sufficient solution but taken in conjunction with all the other - more traditional - identity vetting elements, could significantly complicate the ability of criminals to create (credible) false identities. I’m not saying it would be impossible, but maybe good enough…

  1. parkparadigm posted this