Autonomous Vehicles

Published on February 28th, 2016 | by Kyle Field


Will Self-Driving Vehicles Eliminate The Need For Auto Insurance?

February 28th, 2016 by  

There are over 30,000 deaths each year on the roads of the United States due to operator error… but not for long if self-driving vehicles have anything to say about it. With autonomous vehicles targeting to eliminate 90% of accidents caused by humans, they have the potential to save 27,000 lives per year. That’s 27,000 families that weren’t impacted by a sudden, unplanned, tragic accident.

Looking to accidents as a result of driving under the influence, autonomous vehicles promise to bring major reductions in incidents and related accidents. Nothing different must be done to save lives… the passenger only needs to get in the vehicle as normal and, instead of having to drive home, the vehicle will bring them home safely and — most importantly — without causing harm to others on the road.

The improved safety of autonomous vehicles has been widely discussed, but the massive reduction in traffic accidents related to operator error presents a big opportunity for drivers, as the improved safety should allow them to save money on insurance as well. If accident rates drop across the board by 90%, the need for auto insurance drops accordingly.

What data do we have to support claims that accident rates will go down? Looking at the Google Self-Driving Vehicle program, the cars have put in over 1,000,000 miles on public roads with a total of 17 accidents. That isn’t encouraging until we look at the number caused by the Google-bots… which is zero… and you had better believe that they have the data to prove it!

In parallel to the development of fully autonomous vehicles, “Active Safety” and other “Advanced Driver Assistance Systems” (ADAS) features are rapidly arriving in new vehicles in all tiers — from Nissan, Mercedes, Kia, Tesla… — at all price points. These Active Safety features represent the early stages of the progression towards fully autonomous driving, while also starting to earn the trust of drivers as vehicles take a more active role in driving.

Looking past safety statistics, connected vehicles have the potential to provide extra data to insurance companies to allow safe drivers to get the lowest insurance rates possible.

Stepping back and looking at the insurance industry as a whole, it is obvious that a full frontal assault on the status quo is going to be causing some disruption in the very near future. Will insurance companies follow the many auto manufacturers that were caught flat footed by the electric vehicle revolution, or will they recognize the threat to business as usual and respond with some innovation? Time will tell.

Check out our new 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , , , ,

About the Author

I'm a tech geek passionately in search of actionable ways to reduce the negative impact my life has on the planet, save money and reduce stress. Live intentionally, make conscious decisions, love more, act responsibly, play. The more you know, the less you need. TSLA investor. Tesla referral link:

  • Al

    What does it mean for an automated car in terms of driving that is different from humans

    Rear end collisons

    i) The reaction times of computers are much much faster then that of humans so they will better respond to situations such as sudden traffic stalls. Also humans have been shown to not brake as decisively as needed in these situations probably because you lightly tap the brakes all the time and only on rare occasions you have to floor the brakes.

    ii) If the automated car also has LIDAR or radar it can also see farther than the human eye and therefore better at responding to situations such as the multi car collisions that happen during dense fog.

    iii) Automated cars in conjunction with V2V will work even better once enough cars have V2V capability.

    T bone collisions at traffic intersections due to red light running or cutting across oncoming traffic

    i) These are extremely dangerous collisions since they are side impact which has letter protection then front or rear end collisions. Here an automated car with side sweeping radar is entirely unmatched by humans. Here we just don’t have enough peripheral vision to realize a racing car approaching the intersection.

    Staying in lane or lane merges

    i) Here humans have some advantage currently since computers may not be as good as humans when visually recognizing these. However there is something important about automated cars. Lets say a google car made a mistake in properly recognizing the situation. It most likely won’t result in a head on collision into a barrier due to the safeguards in place but more like a side swipe due to miscalculation. Once this crash is uploaded to google servers that night the machine learning will categorize this situation and learn how to respond to it better the next time. Now most likely the same night an update will be downloaded to every automated car from Google. Now another automated will not make the same mistake. While they will not eliminate accidents they will get exponentially better as more automated cars get on the road. The learnings from each car will help the whole fleet of cars get better.

    Hacking of cars.

    i) Most PCs today have a something call TPM or trusted platform module. The data in the TPM cannot be hacked even if the PC is hacked. Since we are in very early version of automated cars they probably don’t have this. However as they get more mainstream system critical functions could be handled by a TPM like module that cannot be hacked. This will be completely isolated and not interact with any of the connected components that download your email and so on. It is possible design well enough so that even if the car is hacked it won’t result in total control of the car but more like control your interior comforts. I hope the developers of these cars take security seriously so that they keep the trust of people.

  • Brett

    Non-self driving cars can and have been hacked already, that’s not really an autonomous vehicle issue, that’s a connected vehicle issue that will have to be dealt with no matter which direction the automotive industry goes in.

    • Ninjaneerd

      Yup, and right now autonomous vehicles have human back-ups. When this isn’t the case I’m interested to see the direction tech goes to secure against it. Looking forward really 🙂

  • Bob Vittengl

    Yes, the auto industry will change more in the next 5 years , than the past 110 years. Cars will be shared , ins rates crushed , emergency room visits due to car crashes decimited , and gas station will fade away . And the car will act as a back up power source for your home.

    • Jenny Sommer

      On which planet? On earth it will need another 10-25 years to happen.

  • Jenny Sommer

    This part of the cost will just shift somewhat. Maybe you will pay more taxes or buy a software subscription from the car company that will have to pay for insurance anyways.
    Car insurance itself is not really expensive. I pay about 80€/month but more than half of that are taxes. Here we pay taxes for HP…my last car had 160PS and the cost was over 130€/month.
    A self driving car will reduce that by 90% of the pure insurance cost or maybe 30€.
    Here’s the chance to raise taxes by 30€!
    The other savings will go to the car companies that have to provide services for running the autonomous car and constantly improve on the software.
    There will have to be insurance for hacked autonomous cars also.
    Risks will shift but not totally go away.

    • Brett

      Really doubt it. Don’t forget, by reducing automotive accidents and fatalities by close to 90% there would also be substantial savings for healthcare and fire services related to dealing with those accidents. The most expensive and intensive medical care is emergency care.

      More self-driving cars means less need for traffic cops too, most traffic cops don’t generate enough revenue to cover the full cost of their services, so again, another chance to reduce government costs.

      Driver’s licensing bureau? Gone.

      More efficient fleet management reducing traffic will cut down on the need to expand roads and highways too, less capital investment will leave funds available for maintenance of existing roadways.

      Where I come from, snowclearing is a very significant expense, if some of the larger vehicles could be automated, the municipal savings could be very significant.

      The downstream impacts are pretty significant.

      • Jenny Sommer

        Then you are left with new things that have to be payed for. Unemployed Taxi and Transport drivers…
        There is always something to spend money on and since these taxes go into general budget (where I live) they will just ease some spending elsewhere. There are 2b€ needed for refugees, we could need money to ease deficit spending, the universities would gladly take any money, schools have only seen budget cuts the last years.
        Taxes going down because mobility is getting cheaper = snowball in hell.

        • Brett

          Aren’t you just a ray of sunshine.

          Last I checked, global unemployment was less than 10%, while the workforce added about 5.5 billion new workers over the past 100 years. Something tells me the labour market will adapt to the short run impacts of autonomous vehicles.

          • Jenny Sommer

            I don’t doubt the positive effects which are great but the lost tax revenue from insurance/motor taxes and lost fueltaxes with EVs has to be substituted in some way.
            You can’t fire cops or people at the drivers license buero.

            In any case we won’t have to worry about that for at least another 10-20 years and then there will be another 10-?? years transition period.

  • neroden

    No, absolutely not. What a ridiculous question.

    If someone gets killed by a self-driving car, *which will happen*, someone is liable. That person needs insurance. Either it’s the “driver” of the car… or it’s Google and Google needs to buy auto insurance.

    • Kyle Field

      I already responded above but…a 90+% reduction in accidents will shatter the current industry. Redefining who is the driver changes liability. Technology is amazing. The end.

      • JamesWimberley

        SFIK car insurers are not usually completely specialised but divisions of multi-market insurance companies. Risks change all the time. A dramatic and very good reduction in car accidents is part of the day’s work fos a general insurer.

        • Jenny Sommer

          And there are the specialised insurers like the Volkswagen Versicherungsdienst Every 3rd new car from the VW brand (in Europe) is insured with VVD.
          It’s a daughter of VW financial services.

  • Harry Johnson

    Accident avoidance technology will be revolutionary. Accidents just don’t kill 30,000 people a year, they send 2.5 million people to emergency rooms and nearly 200,000 people are hospitalized. Many of these people will suffer from their injuries the rest of their lives.
    But if most of these accidents are avoided, billions spent on hospitals, medical insurance, court cases and auto body repair will also be avoided. This will be a very big change.

    • Kyle Field

      Exactly. Beautiful.

      • nakedChimp

        Hm.. I really wonder when AI will be good enough to do my job.
        I’m very sure we need a new distribution of the earnings-of-production system (and need to get rid of a lot of monopolies, like information, patents, etc. pp), as otherwise this will become the bloodiest and most cruel-some epoch in human history.

      • Raquelwwilson2

        ::f266Work At Home….Special Report….Earn 18k+ per monthfew days ago new McLaren. F1 bought after earning 18,512$,,,this was my previous month’s paycheck ,and-a little over, 17k$ Last month ..3-5 h/r of work a days ..with extra open doors & weekly. paychecks.. it’s realy the easiest work I have ever Do.. I Joined This 7 months ago and now making over 87$, p/h.Learn. More right Here::f266➤➤➤➤➤­nu .❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:❦2:::::::f266…….

  • jburt56

    Not entirely for some time. The problem is that algorithms still have bugs and for quite some time unforeseen situations will arise.

    • Joe Viocoe

      The software is indeed still in beta. But “for some time” is going to be much shorter than people think. The software and sensors can advance with Moore’s Law. Our “trust” in them, will likely be the factor that will take some time.

      • neroden

        The ability of the software to deal with obscure corner cases will always be terrible. So Google Cars *will* kill some people. And someone is going to be liable for that. And whoever’s liable — probably Google — will need insurance.

        The best we can hope for is that insurance rates will go down. The insurance will still be needed. We have very few house fires now, thanks to building codes, but *we still have fire insurance*.

        • Kyle Field

          Agree that the industry will still be there but yes, it will drop significantly. A 90% reduction in risk will shatter the industry as it exists today.

          The fact that Google is also working on legally being defined as the driver is another odd shift. Having a single entity (the manufacturer) being responsible for driving and the liability that goes along with that means cars as a service will make more sense. A shift to fleet driving (uber, Tesla Ride, Lyft, the Googlebot Army etc) makes this essentially a non-issue as the liability for taxis sit with the driver/company, not the passenger. Same idea…new implementation. Should be an extremely interesting 5-10 years in transportation 😀 I’ll bring the popcorn.

          • Jenny Sommer

            Why would anybody need Uber? The insurance cost will be on the manufacturers. Manufacturers would be stupid to sell Uber cars when they can run these services themselves.

            If I remember correctly I have read that Mercedes is planning to run its own fleets once they need no driver.

          • Kyle Field

            Manufacturers will do well but there will still be a place for value add vendors. I envision companies like Uber surviving by doing things like having your lunch ready in the car when it picks you up (and dinner for the family on the way home), picking up and delivering the dry cleaning (when it comes to pick you up and drop you off), delivering personalized media/information/work experiences, integrated showers and bathrooms (why shower before work when you can shower on the way to work?), etc.

            Honestly, I could see Tesla delivering all of these better than anyone else but i have a huge bias towards Tesla because they just keep kicking tail when everyone else (for the most part) appears to be sleeping.

          • Otis11

            Oooh – white parmesan popcorn please!
            (actually, all the specialty popcorns are pretty awesome… I’m not choosy! =-P )

        • Otis11

          Google won’t need insurance… the thing about insurance is you (on average) pay significantly more into it than you get out of it in return – that’s how they make a profit. It’s only beneficial to us as drivers because it allows us to average out the large ‘costs’ that would ruin us individually over lots of drivers for a small ‘fee.’

          If Google owns enough cars to effectively average it out – why would they buy insurance? They’d be better off to put that money aside and just dip into it as necessary…

          • nakedChimp

            Yeah, they would have an insurance dept internally.. that would be tasked to run it efficient.. and before you know it the beancounters want, that they have to make a buck for the Alphabet mother and need a profit, or they will be removed. 😉

          • Otis11

            Google is significantly complex enough that it’s happy to take a hit in one area (All of GoogleX anyone? And Android, etc) in order to help pull in money elsewhere. Plus, it’s negligibly simple to incorporate a slushfund internally to the self driving car department in order to cover insurance related claims. (Actually evaluating the claims won’t be nearly as easy – but paying for them will not be a problem. And yes, I understand the value of money – but they’re going to make a calculated decision on the cost to prevent such accidents, the revenue from rolling out new features, and the cost to pay for the accidents. As such, they will allow accidents and happily pay. The only caveat to this is if it would cause too much public backlash that outweighed the financial aspect.)

Back to Top ↑