Google self-driving car

Insurers ready blame game over self-driving cars

Blogs
Google's self-driving car prototype. (Image: Google)

22 June 2017

Niall Kitson portraitA recent blog post from Allianz chief customer officer Peter Kilcullen hinted at the impact self-driving cars will have on insurance premiums. The simple version says drivers stand to benefit from technologies like LiDar designed to remove human error from the roads. The bad news is that should something goes wrong there will be plenty more blame to go around. Anyone expecting more affordable premiums will be disappointed.

Mind you it will be a long time before we see mass adoptiong of self-driving cars. When they finally come on the market, driverless cars won’t be cheap and could well languish in the leasing market before second and third generation systems force a trickle-down effect to consumers of regular means.

As with the Tesla and the Nissan Leaf, you can expect a lengthy period where the technology will generate more goodwill than sales while regulators  look to use cases and potential implications for speed limits, penalty points systems, etc. I’m sure Minister for Community & Rural Affairs Michael Ring is looking forward to the prospect of Google cars to do regular circuits of remote around the pubs and clubs in an effort to combat drink-driving.

Kilcullen comes at the issue from the insurers perspective, weighing up the respective roles of government, EU regulators, manufacturers, system developers, dealers and drivers – essentially the whole chain from the committee that says you can get behind wheel, to the point when a vehicle makes contact with something it shouldn’t.

“In most European markets vehicle holders are liable for every incident resulting from the vehicles operations,” he writes. “It is mandatory to carry insurance to protect the holder and driver from third party claims, with a primary insurer paying for the claim, regardless of whether it stems from a manufacturing defect or a driver error.”

That’s fine in theory, but what about when the driver is not actually operating the vehicle and is putting their trust in the systems running it, does that mean the developer is at fault?

This argument assumes you can prove who is meant to be in control at the time of incident but Kilcullen has a solution for that too:

“Event data records could come into play when trying to determine who was in control of the vehicle in the case of a self-driving car. However there are legitimate privacy concerns that need to be addressed.

“There needs to be clarity on the data that is captured and how to transfer it in the case of a claim. It’s important to know exactly how data from individual vehicles will be recorded and used to improve safety and clarify liability.

“If a driverless car is involved in an accident liability may not lie with the car owner, but the manufacturer. This is a big change and may cause some traditional personal insurance products to evolve into commercial liability products.”

Data to the rescue once again? In the short term sure, but as we head towards the implementation of the General Data Protection Regulation that clarity over the period of data collection and subsequent deletion may only serve to create ambiguity over control of a vehicle beyond a short timeframe. For example, a driver over the legal alcohol limit takes control after discovering a fault and manages to continue his journey or maybe a driver is distracted from the road and could have been capable of taking over and preventing an incident flagged from some distance ahead. A 15-second window may give the specifics of a collision but not the context.

Some of these questions will be answered by ongoing testing in the US but the EU could add an additional level of complexity over the issue of digital personhood currently being debated.

Back to the short version: If you want to get excited about self-driving cars, you’re more likely to get your fix on YouTube than the M50.

 

Read More:


Back to Top ↑

TechCentral.ie