It’s interesting to see how the nascent Internet of Things is beginning to unfold in the real world.
My fellow Accenture bloggers and I have written a number of posts speculating on the impact that the IoT, including “driverless cars,” will have on insurance. Most of that speculation has centered on the potential our industry can realize once the IoT really begins to take hold.
In a fully autonomous world, things will work very effectively when every device is able to talk to every other device, enabling them to clarify intention and “negotiate” what will happen next.
However, in the meantime, not everything works perfectly.
In a recent incident in Austin, a Google car and a fixed gear bike met at a four-way stop. At the intersection, the cyclist did a “trackstand” – rocking back and forth to keep balance without braking, typical on a fixie. The unfamiliar motion so confused the Google car that it attempted to start and stop, without getting anywhere, for several minutes.
This illustrates the types of risks that are typical in this interim state, where autonomous and conventional vehicles, unable to communicate with each other, share the road. In a fully autonomous world the computer system complexity will be higher, but the interaction risk between vehicles will be much lower. Our near-term challenge is to find ways to integrate unmanned systems with manned systems. This complex “hybrid” environment will represent some new risks.
Perhaps some smart developer (with backing from an insurer?) can create an app that would effectively enable manned vehicles to “broadcast” or behave and negotiate directions like autonomous vehicles – from your smart phone, for example, if you’re on a bike or in an “unconnected” vehicle. This app would effectively put you “on the grid” so autonomous vehicles could “see” you and create more information about what you are and where you’re going. Hopefully this would eliminate some of the embarrassment of other interim challenges.
The current technology of driverless cars is clearly a work in progress. A recent article from Live Science points out several areas where autonomous vehicles can be improved: better software, better sensors, better communication, and – since they have the potential to be involved in life-or-death situations – ethical robots.
But of course the operating rules for autonomous vehicles, even with all these sophisticated inputs, are yet to be defined. And it’s not always clear whose role it is to define those rules. Maybe insurers, based on their role of safely integrating new innovations into society, should be responsible for helping to define those rules.
For more information, go to: