SCU conference on legal issues of robocars


Yesterday I attended the conference on the legal implications of robocars put on by Santa Clara University Law Review. It was a well done conference, with some real content from a varied group of regulators and legal scholars, a sign of how real the robocar world has become.


After a technology introduction where Sven Bieker of Stanford outlined the challenges he saw which put fully autonomous robocars 2 decades away, the first session was on civil liability. The short message was that based on a number of related cases from the past, it will be hard for manufacturers to avoid liability for any safety problems with their robocars, even when the systems were built to provide the highest statistical safety result if it traded off one type of safety for another.

In general when robocars come up as a subject of discussion in web threads, I frequently see "Who will be liable in a crash" as the first question. I think it's a largely unimportant question for two reasons. First of all, when the technology is new, there is no question that any lawsuit over any incident involving the cars will include the vendor as the defendant, in many cases with justifiable reasons, but even if there is no easily seen reason why. So potential vendors can't expect to not plan for liablity.

But most of all, the reality is that in the end, the cost of accidents is borne by car buyers. Normally, they do it by buying insurance. But if the accidents are deemed the fault of the vehicle maker, this cost goes into the price of the car, and is paid for by the vehicle maker's insurance or self-insurance. It's just a question of figuring out how the vehicle buyer will pay, and the market should be capable of that (though see below.)

No, the big question in my mind is whether the liabilty assigned in any lawsuit will be significantly greater than it is in ordinary collisions where human error is at fault, because of punitive damages. The cost of collisions is well established and understood, and there is a large industry to manage it, and if robocars are, as planned, safer, then that cost goes down, and that means savings for the car buyer and for society -- a big win for all. However, if the cost per collision is much higher even though the number of collisions drops, this can impede the ability of a robocar industry to innovate or even to exist and save those injuries.

Unfortunately, some liability history points to the latter scenario, though it is possible for statutes to modify this.


All cars must have insurance today, and it is normally bought by the car owner/driver, and covers almost all accidents. If a safe robocar is delivered, the lower rate of accidents should mean cheaper insurance. Alas, in California, one of the world's largest insurance markets and home to robocar labs at Google, Stanford/VW and others, it's not so simple. California's Proposition 103 demanded that any insurance policy's price must be based on weighted factors, and the top 3 weighted factors must be, in order, driving record, number of miles driven and number of years of experience. Other factors like the type of car you have -- ie. if you have a robocar -- must be weighted lower. So this law makes it very hard to give very cheap insurance for a robocar, and makes it close to impossible to do it for somebody who is getting the robocar precisely because they are a bad driver and have found themselves facing expensive insurance if they keep driving.

Because Prop 103 is a ballot proposition, it can't easily be superseded by the legislature. It takes a 2/3 vote and a court agreeing the change matches the intent of the original ballot proposition. One would hope the courts would agree that cheaper insurance to encourage safer cars would match the voter intent, but this is a challenge.


Kevin Vincent and Steve Wood from NHTSA, which writes the federal motor vehicle safety standards and does the crash tests, among other things, spoke and gave a fairly robocar-positive message. They are working to assure the vehicles operate safely on the roads, but do appear to be fully aware of how they are a new technology under new rules, and they don't want to impede adoption by overregulating. They are working out plans to test vehicles in simulation, on test tracks and on the road to rate for safety and seem to have a generally forward-thinking plan.

Local and criminal laws

The session on criminal laws centered more on the traffic code (which isn't really criminal law) and the fact it varies a lot from state to state. Indeed, any robocar that wants to operate in multiple states will have to deal with this, though fortunately there is a federal standard on traffic controls (signs and lights) to rely on. Some global standards are a concern -- the Geneva convention on traffic laws requires every car has a driver who is in control of the vehicle. However, I think that governments will be able to quickly see -- if they want to -- that these are laws in need of updating. Some precedent in drunk driving can create problems -- people have been convicted of DUI for being in their car, drunk, with the keys in their pocket, because they had clear intent to drive drunk. However, one would hope the posession of a robocar (of the sort that does not need human manual driving) would express an entirely different intent to the law. While the car might have an emergency manual override, as is likely for other reasons, one would hope it is the drunk's responsiblity if they use it, not the car's responsiblity for having it there.


The session on privacy left me a little wanting. There was no addressing the issue that cars will have cameras on them, recording the whole world once they get frequent enough. Most of the concern was on the logs of all your movements, which is indeed a concern -- and already one with cell phones.

ITS and Spectrum

There were sessions on ITS and DSRC (the spectrum allocated for communications from vehicles to other vehicles and to local infrastructure points.) I've written on this topic before, and while I believe that of course robocars will make use of any such signals, they can't be built to depend on them, and will in fact only encounter them occasionally for the next decade at least. As such, robocar development proceeds without much connection to the connected vehicle world, which I am sure bemuses those in the connected vehicle world.

You can read another press report on the conference and one from John Markoff of the NY Times.


Thank you for this report. The social issues are at least half the battle.

On liability for crashes, a lot will hinge on the expectations of human parties involved. Consider some scenarios where robocars are mixed with human drivers.

1: The car was doing the right thing to avoid a crash but the human panicked, overrode control, and made things worse. The driver claims the car failed to inform him that actually it had things under control.

2: The car is driven offroad or outside its realm of operation but is then called on to function, but fails in some way. The driver says his other robocar has that capability so he expected it in this model.

3: The car notifies the driver that it is not certified to operate in heavy rain at night. The driver knows that he is clueless under these conditions and makes the car drive anyway. The car cannot say no, but it slows down a lot which causes a chain pileup behind.

4. Human drivers learn that robocars are cautious and they develop strategies to cut in front and otherwise take advantage of them. The human driver of the robocar steps on the gas to prevent this cheating and gets blamed for overriding unsafely.

5. While cruising down the highway, the lidar gets hit by a rock and notifies the driver that a technical problem has occurred and he should take over. The human was sleeping even though they are instructed to be ready to drive if necessary. Nothing had ever gone wrong in thousands of hours of robocar driving so he claims he was conditioned to never take control.

A lot of law is based on what a reasonable person would be expected to do. Part of the solution has to be massive public education campaigns to give everyone a common understanding of what robocars are about. This will be difficult in part because the technology will be a moving target for a long time. The wrong movie could lodge boneheaded ideas in a lot of peoples' minds.

You point out that fundamentally, if robocars provide huge value, the legal and insurance systems should be able to figure it out. The problem with value coming from safety is that its distributed unevenly, and its out of sight. You don't know about the accidents that didn't happen. The problem with value coming from reduced congestion is that this effect only kicks in after lots of people have robocars. Early adopters don't benefit. This suggests that at least in the early stages, the value proposition needs to be readily visible, perhaps about use of time. A second high-visibility selling point could be about being able to use your car legally after getting plastered, but this can only happen with more advanced robocars that support driving without human backup.

Scenario development is a specialty of science fiction authors. Has Google considered sponsoring a science fiction contest with a robocar theme?

Yes, just as you say, there will be things where the liability goes squarely on the driver/owner and other times when the liability will go squarely on the vendor. And in many cases there will be some debate about who it is.

This brings up something I didn't discuss -- it could be that this very debate is expensive, to both sides. As I have said, what matters most is if the total cost goes up so much that the technology is not economical. But this is not greatly different from the problem faced thousands of times a day when two humans have a crash, figuring out who is at fault, whose insurance will pay. Insurance companies know there is no point in wasting a lot of money on this most of the time, and in some jurisdictions have even got no-fault laws in place.

I've predicted the more complete robocar will come with insurance. This insurance will cover both robocar operations and manual operations. It does not matter that the safety risks are not evenly distributed -- this is what the insurance industry is all about solving. They hire tons of actuaries who are well skilled at working this out. The insurance companies, and their customers, care only about the total insurance cost which is then shared among all policyholders, distributed according to general risk factors.

There, things like Prop 103 could screw up the mix.

(Sorry Brad if my previous comment broke your rule about Homepage links).

Regarding this statement:

"the problem with value coming from reduced congestion is that this effect only kicks in after lots of people have robocars. Early adopters don't benefit. This suggests that at least in the early stages, the value proposition needs to be readily visible, perhaps about use of time."

Obviously the value from reduced congestion is going to take a while to play out, but there will be a lot of other value offered which will more than adequately compensate for the wait:

- being able to avoid parking and the associated costs
- cheaper travel as the car will almost definitely be 100% electric
- the economics of mass-driverless-car ownership by certain operators (Zipcar et al) will mean the ongoing cost of owning your own car will be unjustifiable in the face of the new situation.

As insightful as ever. I'd like to weigh in here on a couple of specific points:
Cost of liability- we tend to look at the cost of insurance in terms of injuries and repairs. Another significant layer is the cost to defend. This will be particularly messy during the early adoption phase. Today, a claim is against a driver who made a mistake. With these vehicles the range will run from the driver, to the manufacturer, the software engineer, the chip maker, the manufacture of each component, the wireless carrier, and so on. The litigious nature of America may well prevent lives from being saved.
Next, to complicate matters more, the bulk of our liability system is regulated differently across all 50 states.
Finally- let's remember that we that the stats can be put into a story to highlight why working through these issues will happen: imagine consumers spending 4 billion hours a year, waiting to board 747's, knowing in the US, 1 will crash every 4 days. Who would do it? On a global basis, it's a 747 every 24 hours. These debates remind of others like why entertainment material can never be streamed due to piracy- whoops. Or- why genetic scanning won't lead to preventative health care because of cost and privacy...please...really?
In the end, it is us-consumers that will set the pace. If we want to stay connected and entertained, we'll have to let the car drive. Take control and all that fun, extra productivity, social interaction will just shut off. What do you think we will choose? PLEASE send me my robo taxi!

Unlike most systems, software systems are deeply instrumented and logged. While there are a number of privacy concerns about this, I actually expect there will be very little to debate about the cause of most robocar accidents. That won't stop people trying to debate, I agreee. But after a time, with a complete 3-D video of the accident and detailed logs of the computer's internals, the defendants will know when to fight and not to, and the courts will tire of crazy tangents. But at the start, it's harder to predict, and punative damages are the issue. But we won't be debating if it was a bad chip at fault; I suspect we'll know quite clearly.

Hi Brad

First off, I love the site! I am currently putting together a blog to track worldwide Driverless Car news so you'll see me hanging around here a bit more leaving comments.

Most of the speculation around this topic appears to me to be based on false assumptions - that the current model of car ownership will persist.

I would suggest that over 90% of robocars will not have individual owners - they will be owned by operators functioning much like a taxi service of today - or as evolved versions of what Zipcar is today.

(It makes much more sense commercially. Taxi services are expensive today for two broad reasons, those being the person driving and the cost of fuel. The implications of eliminating both of those two costs is mind-boggling. Imagine what would happen if taxis in the city where you lived suddenly dropped by 70% in price?))

This will probably mean that the consumer-side legal issues will have to be dealt with the by the operator of the service and their insurer.

Well, I might suggest there already is a fine blog with worldwide driverless car news, but of course there can always be more than one. :-)

The truth is I don't think we know how the car ownership balance will settle out. There will be privately owned cars never used by other than the owners and their friends. There will be taxi fleets owned by taxi companies. There will be private cars hired out as taxis either directly or through brokers. How this will actually balance will depend on the quality of the product and the demands of the market. I've spoken to too many people to believe we can predict this today. For example, in New York City, where car ownership is expensive and parking is difficult and taxis are plentiful, there is still 25% car ownership. Robocars make robotaxi service easy and cheap, but they also make owning easy and cheap because you don't worry about the parking and fueling, the car does.

The cost of taxis ($2.50 to $3 per mile) is not really gasoline -- 7 cents/mile in a hybrid taxi, 20 cents/mile in a guzzler. The person is a larger cost -- at 12 miles/hour, that's $30/hour and the hack probably makes $8 to $10. A lot of it is captured by the medallion owner, which is often not the driver. Fortunately robocars should help us eliminate the medallion system. Though not without a big political fight.

Hi Brad

"Robocars make robotaxi service easy and cheap, but they also make owning easy and cheap because you don’t worry about the parking and fueling, the car does."

Sort of agree with this, sort of don't. Yes, it does drop the ongoing cost, but we also have the issue of vehicle financing which forms a large part of the cost of owning a vehicle. Due to economies of scale, my hope/prediction is robotaxi companies will engage in mass-buying of driverless cars and build instantly profitable fleets even while the prospect of autonomous driving is still in its early days in terms of private ownership. Assuming $150k per car, profitability would be easy given the competition at first won't be private ownership but taxi operators.

In terms of the medallion system, I also (sadly) can't imagine us being able to avoid the big fight, especially given the mafia-like tendencies of so many taxi operators. The extent of job loss will be horrendous.

"The cost of taxis ($2.50 to $3 per mile) is not really gasoline — 7 cents/mile in a hybrid taxi, 20 cents/mile in a guzzler. The person is a larger cost — at 12 miles/hour, that’s $30/hour and the hack probably makes $8 to $10"

Ha. Ok. This completely depends on the city - I didn't have New York or comparable cities in mind when I posted. In my city the numbers are different - I don't have the data, but in Melbourne our taxis drive a lot of kilometres per shift which means that the cost of petrol almost equals that of the driver. It's a regular occurrence for the taxi to spend more on petrol than the driver nets. (Just from anecdotal evidence).

Anyway, I have spent the last few hours reading through the essays of your site and agree with pretty much all of it. I have been brainstorming this stuff out and the logical conclusions (especially once driverless bots are operational) are mind-blowing.

The one thing I hadn't considered was short-term rental of hard-goods.. which of course makes perfect sense.

One thing before I go - myself and my blogging partner have spent a bit of time lately discussing the impact that this will all have on organ donation. It will kill organ transplants dead cold unless we can figure out ways to manufacture the organs.. or even go down the path of compensated organ donorship.

Oh, taking a taxi will be cheaper than ownership, but we're yet to see how the market will decide between the two. With electric vehicles, the cost of the electricity is quite low (penny per mile or less) and becomes unimportant. Depreciation (including on the battery) is the big factor.

When the market moves to having far more taxis, you also get a more serious effort to design vehicles to be a taxi. Today they are minor modifications of existing cars. For a taxi you want:

  • High duty cycle. Run all the time for 3 years than die, rather than run intermittently for 15 years then die.
  • Possibly amenities like electric doors
  • The camera system I describe which photographs the interior before and after the occupant is present, to see if there are differences, ranging from forgotten items to spills.

And some others. The high duty cycle is a challenge for electric vehicles, though. You may want to talk battery swap on taxis where private vehicles don't really need it. Or if the battery is 70% of the cost of the vehicle, you might just build lots of extra taxis rather than swap batteries.

The taxicab companies will be thrilled to buy robocars. The drivers will be upset. In some towns the drivers are the owners -- some towns require driver ownership of medallions at some level. The medallion owners will fight tooth and nail anything that breaks their monopoly. In most cities though, including NYC, the medallion is a monopoly on picking people up who hail you on the street. So private cabs hailed from cell phones, for now, are allowed.

But nonetheless people like car ownership. They like expressing themselves with their car. They like choosing their car. They like to leave their stuff. I have solutions to most of these, but I still think there will be lots of owners, at least for a generation.

Regarding the solutions, I did see your "stuff box" idea. This is interesting and will maybe end up being a point of difference for one particular robotaxi operator?

It will be a fun exercise watching them differentiate - luxury, waiting time, cost, features... there will be lots of ways. Maybe one brand will be completely funded by advertising? (Watch an ad at the start of the trip, ride the first 5 minutes for free?).

I think the world will most likely just adopt a New-York attitude to carrying things around.

If you see the article on my blog today, Better Place has just delivered their first cars to Israel. Their battery-swap technology will surely be the most viable solution.

If the stuff locker plan is to work, you need some reasonably standardized sizes, not just among the Taxis but with the private cars. One reason private car ownership is made easier by robocars is you no longer need to buy a car that meets all your needs because you can hire one for special needs -- cargo hauling, 4WD trip, large group, longer range, your battery is almost out, couple with only 1 car sometimes needs 2, etc. If you no longer need a car to handle all those things you might be happy with a much simpler (and cheaper) 1-2 person car for example. But then you need to transfer stuff easily, so standardized lockers are a win.

I also think BetterPlace is not a great idea for private cars unless you have a lot of them around, but it's great for robocars and taxis, which don't mind returning to a depot as long as it's not too far away to do a battery swap. BetterPlace needs cars to all standardize their battery shape and also for car owners to not own their batteries. However, a taxi fleet can get by with just a few swap stations and it does not need to be compatible with other cars.

EN-V Cars or LIT cars are how I envisage most personal transport to happen, given that such a high proportion of day to day journeys are conducted solo.

You wouldn't happen to know of any efforts to standardise battery design or is it too early?

It should be interesting with Better Place to see how taxi companies react as this will be a good early signal of how things will take off in Driverless Car land.

On another topic, do you have any thoughts on the CAFE Foundation Challenge & Pipistrel USA's efforts? (for those reading -

It seems like this has almost immediate potential for pilot-less development considering that a) auto-pilot technology is already pretty much good enough and b) the quantity of potential hazards is much smaller. It seems like the biggest obstacle would just be a lack of a sufficiently motivated big-money company.

I did attend the Green Flight Challenge day at NASA, so I know about it, and automated flight is interesting but is a different problem. However, you can see my notes on airports, "flying cars" (non-standard meanings of that phrase) and others on this blog. You still need robocars to get to the airports, and of course a new air traffic control system -- which was built but not deployed at NASA.

Better Place is a tough problem. Being able to swap batteries sure would be nice, but so much stands in the way. I think that the industry is too young to standardize the battery pack. Right now a lot of the innovation in electric cars is taking place in the battery pack, and I expect this to continue for some time. With Better Place, customers are not buying the battery, and it must fit a standard form factor. The first is just as dangerous to innovation -- you want products that are bought by early adopters, one at a time, for high speed innovation. One size does not fit all when it comes to electric car batteries.

And the other hard part is that it's a lot of work to get the swap stations in so many places that it really delivers the convenience people want -- for human driving. For robocars, the swap station model works fine because you can have different swap stations for different battery cases, and they do not have to be super conveniently located, and robots don't even mind a 15 minute wait for a swap (or they make an appointment for one.)

Thanks Brad. Should make for an watch.

While the Western democracies might take a while to figure this out, the more autocratic governments of China, Singapore or Taiwan or Korea may just mandate an insurance scheme simply to solve their immediate problems (traffic accidents really couldn't get much worse in China) and at the same time position themselves in the fore front of providing a battle tested solution to the Western world shortly thereafter.

With luck perhaps one or more of the major car companies in the West may convince things to move faster here, but I'm not holding my breath. But even if it happens first in China that won't delay it happening here by more than a few years. Once (we'll assume there are some) the benefits are obvious things will move with Internet speed to get adopted widely.

Really the big problems are the ones alluded to above. Many entrenched interests will fight tooth and nail to maintain their monopoly over various parts of the transportation system. And to do that will throw pretty much any and all types of arguments out to see what sticks.

Yes, I've written a number of times that the liability system may be much more favorable in some countries.

On the other hand, the driving in China does make the problem a lot more difficult to solve. India is even worse. In China the line down the middle of the road is just as suggestion. In India the direction of the lane is just a suggestion.

My hope is that we get one of those rare enlightened moments from the government that got us useful things like the safe harbours of the CDA and DMCA. In fact, a good project in China or Singapore would actually demonstrate that US jurisprudence is creating a risk that the US will be the follower in this important technology though it starts out as the leader, and perhaps action would be taken to correct that.

"It was a well done conference"

But how was the bag?

It was a free conference for academics, cheap for business, so pleasantly it did not have a bag. The schwag was a pen.

To be fair, I didn't write this joke. I think I first heard it from Steven Schladover at UC Berkeley, who hates that people call them autonomous cars (the thinks it should be "automatic") and is in general a skeptic about robocars on ordinary highways even though he's worked on things like platooning and cars that follow magnets.

The reason for the word autonomous comes form typical vocabulary in robotics. At first, robots were really just remote controlled. As people built robots able to make movement decisions on their own, not just on how to get from A to B, but doing dynamic things like avoiding moving obstacles etc. this got called autonomous because these small movement decisions are autonomous, even though the overall goals are not. Steven won't win his fight, but his joke is correct. A fully autonomous vehicle would be one that could do what it wanted to do in spite of instructions. Of course, for now, what it "wants" would be the result of other programming.

Add new comment