Would we ever ban human driving?


I often see the suggestion that as Robocars get better, eventually humans will be forbidden from driving, or strongly discouraged through taxes or high insurance charges. Many people think that might happen fairly soon.

It's easy to see why, as human drivers kill 1.2 million people around the world every year, and injure many millions more. If we get a technology that does much better, would we not want to forbid the crazy risk of driving? It is one of the most dangerous things we commonly do, perhaps only second to smoking.

Even if this is going to happen, it won't happen soon. While my own personal prediction is that robocars will gain market share very quickly -- more like the iPhone than like traditional automotive technologies -- there will still be lots of old-style cars around for many decades to come, and lots of old-style people. History shows we're very reluctant to forbid old technologies. Instead we grandfather in the old technologies. You can still drive the cars of long ago, if you have one, even though they are horribly unsafe death traps by today's standards, and gross polluters as well. Society is comfortable that as market forces cause the numbers of old vehicles to dwindle, this is sufficient to attain the social goals.

There are occasional exceptions, though usually only if they are easy to do. You do have to install seatbelts in a classic car that doesn't have them, as well as turn signals and the other trappings of being street legal.

While I often talk about the horrible death toll, and how bad human drivers are, the reality is that this is an aggregation over a lot of people. A very large number of people will never have an accident in their lives, let alone one with major injuries or death. That's a good thing! The average person probably drives around 600,000 miles in a lifetime in the USA. There is an accident for every 250,000 miles, but these are not evenly distributed. Some people have 4 or 5 accidents, and many have none.

As such, forbidding driving would be a presumption of guilt where most are innocent, and tough call from a political standpoint.

That doesn't mean other factors won't strongly discourage driving. You'll still need a licence after all, and that licence might get harder and harder to get. The USA is one of the most lax places in the world. Many other countries have much more stringent driving tests. The ready availability of robotaxis will mean that many people just never go through the hassle of getting a licence, seeing no great need. Old people, who currently fight efforts to take away their licences, will not have the need to fight so hard.

Insurance goes down, not up

You will also need insurance. Today we pay about 6 cents/mile on average for insurance. Those riding in safe robocars might find that cost down to a penny/mile, which would be a huge win. But the cost for those who insist to drive is not going to go up because of robocars, unless you believe the highly unlikely proposition that the dwindling number of humans will cause more or deadlier accidents per person in the future. People tolerate that 6 cent/mile cost today, and they'll tolerate it in the future if they want to. The cost will probably even drop a bit, because human driven cars will have robocar technologies and better passive safety (crumple zones) that make them much safer, even with a human at the wheel. Indeed, we may see many cars which are human driven but "very hard" to crash by mistake.

The relative cost of insurance will be higher, which may dissuade some folks. If you are told, "This trip will cost $6 if you ride, and $8 if you insist on driving" you might decide not to drive because 33% more cost seems ridiculous -- even though today you are paying more for that cost on an absolute scale.

Highly congested cities will take steps against car ownership, and possibly driving. In Singapore, for example, you can't have a car unless you buy a very expensive certificate at auction -- these certificates cost as much as $100,000 for ten years. You have to really want a private car in Singapore, but still many people do.

Governments won't have a great incentive to forbid driving but they might see it as a way to reduce congestion. Once robocars are packing themselves more tightly on the roads, they will want to give the human driven cars a wider berth, because they are less predictable. As such, the human driver takes up more road space. They also do more irrational things (like slow down to look at an accident.) One can imagine charges placed on human drivers for the extra road congestion they cause, and that might take people out of the driver's seat.

The all-robocar lane tricks

There are certain functions which only work or only work well if all cars are robocars. They will be attractive, to be sure, but will they surpass the pressure from the human lobby?

  • It's possible to build dynamic intersections without traffic lights or bridges if all cars are trusted robocars.
  • It's possible to build low-use roads that are just two strips of concrete (like rails) if only robocars go on them, which is much cheaper.
  • It's possible to safety redirect individual lanes on roads, without need for barriers, if all cars in the boundary lanes are robocars. Humans can still drive in the non-boundary lanes pretty safely.
  • We can probably cut congestion a lot in the all-robocar world, but we still cut it plenty as penetration increases over time.

These are nice, but really only a few really good things depend on the all-robocar world. Which is a good thing, because we would never get the cars if the benefits required universal adoption.

But don't have an accident...

All of this is for ordinary drivers who are free of accidents and tickets. This might all change if you have an accident or get lots of tickets. Just as you can lose you licence to a DUI, I can foresee a system where people lose their licence on their first accident, or certainly on their second. Or their first DUI or certain major tickets. In that world, people will actually drive with much more caution, having their licence at stake for any serious mistake. A teen who causes an accident may find they have to wait several years to re-try getting a licence. It's also possible that governments would raise the driving age to 18 or 21 to get people past the reckless part of their lives, but that this would not be a burden in a robocar world, with teens who are not even really aware of what they are missing.

I've driven over 35 years and had no accidents. I've gotten 2 minor speeding tickets, back in the 80s -- though I actually speed quite commonly, like everybody else. It seems unlikely there would be cause to forbid me to drive, even in a mostly robocar world. Should I wish it. I don't actually wish it, not on city streets. I still will enjoy driving on certain roads I would consider "fun to drive" in the mountains or by the coast. It's also fun to go to a track and go beyond even today's street rules. I don't see that going away.


You mentioned congestion-based disincentives to human driving, but you've left out dedicated rights-of-way. This can start by giving robocars unrestricted right to use existing restricted or semi-restricted rights-of-way, such as diamond lanes, BRT lanes, and bus/taxi-only lanes. Later this can expand to more lanes and/or entire streets that human-driven traffic is restricted from using.

It's also easy to imagine certain rights-of-way that (non-emergency) vehicular traffic is currently restricted from entering due to safety concerns that might actually be opened up to robocars. This is especially applicable for smaller vehicles and/or ones flagged as in-use as paratransit. This would be similar to the small electric vehicles that are occasionally used in airports.

"Today we pay about 6 cents/mile on average for insurance. Those riding in safe robocars might find that cost down to a penny/mile, which would be a huge win. But the cost for those who insist to drive is not going to go up because of robocars."

Insurance companies seem to seek out any excuse they can manage for raising insurance rates.

At the very least, they will grandfather in existing plans, and any new subscribers or people who change their plans will be forced to accept the new, higher rates.

When people start shifting to robocar usage and fewer people get driver's licences at all the auto insurance companies are going to start feeling a crunch as their subscribers bases shrink and revenues go down. They might lower their rates for perfect drivers to incentivize them to keep driving at all, and impose more and more penalties on imperfect drivers to make up for lost revenue.

As for raising the driving age, I think that'll shake itself out on its own and it won't be necessary. Already fewer and fewer young people are getting licences and cars. With robocars available this will probably accelerate and it'll just become a moot point. Plus people in rural areas at least will strongly resist raising the driving age.

In a similar way, as safety goes up and accidents go down with more robocars on the road, there might not be many calls for increased licence revocation based on driving history.

Beyond all this, there are tons and tons of things that humans do that are dangerous to themselves and/or others around them, and yet they're extremely resistant to have these things restricted due to their "rights," especially in the U.S.A.

Better just to have usage- and cost-based incentives for better behavior/technology.

You would be surprised, they are not nearly so money grubbing as you think. First of all, several of the largest insurance companies are mutual insurance companies who distribute profits to policyholders.

But to my surprise, I learned that most insurers actually plan for a loss every year. They make their money not on profits on the premiums, but by investing the float. Americans hand $200billion over to insurance companies at the start of the year, which they pay out slowly over the course of the year. Even if they pay out $201B, they make quite decent money, because they are really investment companies inside.

The truth is that premiums = (cost per accident * number of accidents + operating costs of insurance co)/ number of policyholders, with minor adjustments for operating expenses etc. For premiums to go up:

  • The cost per accident might go up. Note that most accidents do not involve courts or juries, but just use well established numbers. To go to court easily surpasses those numbers and eats up all value -- most of the time.
  • Number of accidents by humans will not go up, though it could be that the self-drivers are a riskier than average bunch? I see no big reason for that.

It's really in the (number of accidents / number of policyholders) figure.

People who opt to only use robotaxis instead of owning a car will reduce the total number of policyholders. This would lead to an overall increase in premiums.

Car owners can then be subdivided into two groups: Self-drivers and robocar owners.

If, as we generally agree, a person in a robocar is less likely to have an accident than a self-driver, then as a sub-group, the self-drivers will have more accidents per policyholder than the robocar owners. By your calculation this self-driving sub-group then should see higher premiums than the robocar owners.

It could be argued that this difference will be seen by the premiums of the robocar owners going down, but then that would mean that the total revenue of the insurance companies will go down. Instead, it's more likely that the premiums for robocar owners would drop a little, but those on the more at-risk group (the self-drivers) will go up.

"Robocars will result in insurance premiums on non-robocars going up" is not a great sales pitch for robocar acceptance by the general public, though. ;)

Many states are actually pushing to require insurance be charged per mile, and it's a good way to think about it. Today the typical cost is 6 cents/mile.

Robocar operation will have a very low accident rate per mile, perhaps just 1 cent/mile, paid by the car maker or operator, perhaps.

Self-drive operation will have a higher cost per mile. As long as the self-drivers are a representative subset of today's drivers, it will still be 6 cents/mile. Actually probably less, because tomorrow's cars will be safer, including many accident avoiding technologies, even when human driven. So perhaps 4 cents/mile.

It only goes up if the subset of drivers who self-drive are weighted towards the less safe drivers today. I actually think that's wrong, I think it will be the other way, because I hope we'll be taking the drunks out of the equation -- almost all of them not driving, and so, as a group, these drivers will have a lower accident rate per mile, and thus lower insurance than today.

This is for liability and collision insurance. Comprehensive remains the same, pretty much.

But that all means that total pool of revenue for the insurance companies goes down. So how are they likely to react to that?

They already know that's coming. But they plan to just adapt to it. Just as they don't oppose flame-retardant construction or anti-theft systems, which also reduce claims and revenue, when they work.

They will look for other things to insure. Some will not adapt well and die. It's the way of things.

I suspect banning human drivers will happen faster than that.

Think in terms of MADD (Mothers Against Drunk Drivers.) They lobbied for years to eliminate drunk driving to reduce drastically the number of people killed by drunk drivers.

As soon as robocars are shown to be significantly safer than regular (non-drunk) drivers it is possible (probable?) that they will retarget their efforts against human drivers as the number of people being killed by human (non-drunk) is larger than what they targeted drunk drivers for.

So possibly MAHD (Mothers Against Human Drivers.) And while it took several decades for the original campaign to take full effect any new campaign will benefit from their expertise, experience and social media.

It is totally unacceptable socially to drive drunk today (unlike it was back in the fifties and sixties.)

I think it will be likewise socially unacceptable to drive yourself in the not too distant future. Possibly within five years of when robocars are easily available. Ten years at the most.

They are not that successful. The USA, with its common .08 limit, is one of the most tolerant nations in the world on drinking.

However, one valid point is that as robotaxi service becomes available, the limit could drop, and even to close to zero. In other words, if you have anything to drink within the last few hours, take the taxi, or have your own car drive you home.

But that's not banning human driving, it's banning being stupid.

Yes human drivers will be banned...for some time(first couple of years)they will pay 100k $ insurance if they want to drive.

I don't think cars will ever be fully automated unless there is a way for law enforcement to monitor the vehicle or be able to pull it over and search it. As of now, LE makes up silly excuses to pull people over and search their cars (drug interdiction program run by the DEA). Without some sort of ability to search cars traveling across the country, I doubt there will ever be fully automated highway vehicles.

You are correct that police will have a great deal more trouble pulling over a person in a robocar, both because they will not have much to use as grounds for suspicion, and because if they lie, the car will likely have recorded that there were no grounds.

I don't see the police stopping the technology because of that, though,

However, since on our roads today, most people speed, they will also desire to speed in their robocars, and that will give the police cause to pull you over. I believe the speed limits should be adjusted.

Good thoughts in this article; I've also been idly pondering this issue of banning drivers. I'm also someone who loves to drive, is one of the <5% who has a stick-shift in my daily driver, has done quite a bit of amateur sportscar and ice racing. Yet I very much look forward to automated cars because daily driving rarely fun anymore with the current levels of traffic. I just want to get in the car and have it get me there while I work or relax. So, I expect that others who care about driving much less will gravitate away from manually driving even more readily, so those of us that like to drive will likely be a rapidly shrinking minority, with commensurately small political clout.

While I agree with the strong inclination and history of grandfathering rights, I really wonder about the networking and swarm capabilities of automated cars. This will enable 1) highway convoys bump-drafting with 6-inch gaps with huge fuel savings and increased road utilization as well as 2) traffic-light-free intersections with cars interlacing their turning paths with 6-inch gaps like high-speed loom shuttles. For #1, the only humans that could participate in the highway convoys would be highly trained pro race drivers who are particularly alert that day. For #2, a single human would cause a massive mess at any automated intersection, either forcing a major back-up as all the cars avoided the clumsy human, or a big crash as they failed to avoid the outlier.

The simplistic way that most govt's deal with things would be to just ban humans from at least those situations, so it seems that banning humans is a real likelihood.

The only way out of it that I see is if the human driven cars also get the full sensor and comms packages that allow full networked comms with the other cars so that they can automatically avoid the humans, or perhaps that cause the car to just take over in certain zones.

I'd love to be wrong on this -- does anyone see a way that the technology trends don't point in this direction?

The fuel savings from close spacing are modest (15% or so) but still worthwhile. They don't require banning human drivers. Actually, they don't even require the lead car to be a robocar, really.

The smart intersections require all-robocar, but I don't see this happening for old intersections for decades because of the reasons outlined. Robocars will be able to do v2v -- very old cars will not, but the first cars have nobody to talk to so this stuff happens later.

Close spacing at higher speeds require cleaner, better maintained roads than we have at present.

I fear that the possibilities and benefits of platooning for robocars are overrated and thus overemphasized.

I think the close spacing abilities of robocars are more practical in urban environments, as described in the OP as a means to improve and smooth traffic flow at intersections. That'll require either restricted rights-of-way that human driven vehicles are not allowed to enter, or perhaps triple-mode intersections: 1) All cars (human driven and robocars) in one direction, 2) All cars (human driven and robocars) in the other direction, 3) Robocars only in all directions. Versions of this already exist at some intersections for "all way" pedestrian crossings and transit vehicle priority, especially where there are dedicated transit rights-of-way (San Francisco has this at some intersections already, and it's planned to be implemented more in the near future for bus rapid transit on Geary and Van Ness).

Yes, you may well have read those facts about platooning here. They are not widely known.

However, there is a difference between platooning close enough to draft and save fuel, and dropping to short headways like 1 second or even half a second. Typical platooning distance is 3 to 5 meters, which is 0.1 to 0.2 headway at 70mph -- though more correctly with 5m long vehicles it's about 0.3 seconds per vehicle. 0.5 seconds is about 16m per vehicle, which is not really platooning, but approaching it.

Anyway, even 1s spacing almost doubles lane capacity. Of course, you don't get double because all vehicles are not robocars. But a robocar will eventually be able to follow any vehicle (human or software) at 1s, while human drivers average just under 2s. In theory they might follow just a bit closer on cars doing V2V, which will transmit brake hits and other data to you, but your radar doppler tells you quite quickly when the guy in front of you changes velocity.

There is risk to this (certainly at 0.5s) because there's more than can go wrong if things go wrong.

I read a study about platooning elsewhere, and I know I've commented and we've discussed it here before. :)

"Anyway, even 1s spacing almost doubles lane capacity"

...at freeway speeds. But greater capacity is more highly desired during rush hours, when roads are actually over capacity and the traffic slows to a crawl.

When the traffic is already flowing at full speed a road does not actually need more capacity. Platooning can enable full speed travel for a greater capacity of vehicles, but there tends to only be a short window of time when a high speed road goes from full capacity at full speed, to double that capacity, to exceeding it. So this kind of platooning does have some benefit in this regard, but I think it's easy to over-exaggerate it. My wild-assed-guess is that the benefit would exist for only maybe 10-15 minutes near the start and end of rush hours.

Plus there's the problem that increasing a road's capacity in any way, such as adding a lane, only actually helps temporarily, as that extra capacity quickly gets taken advantage of by new traffic that might not otherwise have been using that road.

Yes, congestion comes when demand > capacity. But if you take a road that's becoming stop and go and double the capacity of the lanes it will open up.

Induced demand is an issue -- though serious increases in capacity give you a large buffer zone. However, metering is the answer to induced demand. Don't let more cars on the road than it can handle, and you will get far less congestion.

if you take a road that’s becoming stop and go and double the capacity of the lanes it will open up.

Again, though, that's only for a short window until the road gets over capacity, which is the bulk of the time at rush hours.

However, metering is the answer to induced demand. Don’t let more cars on the road than it can handle, and you will get far less congestion.

Of course, this is wholly independent of self-driving vehicles. It doesn't even need V2V or V2I, it only needs sensors on the road and a system of "smart meters" at the onramps that get information from the sensors and talk to each other in a coordinated way. Adding V2I gets you in-car navigation systems that give better routing to both human-driven and self-driving vehicles.

There's only so much that can be done to siphon off traffic onto surface streets or alternate routes when there's simply too many vehicles for the whole road infrastructure to support them all traveling at full speed. In the end you're still going to have a significantly longer trip duration during peak times. Either you'll wind up sitting, waiting in your vehicle somewhere, either in stop and go traffic on the freeway or at a metered onramp, or you'll be on slower surface streets, dealing with signal lights, stop signs, etc.

The greater benefit to having robocars during these times is that you'll be able to spend that extended time while you're in the vehicle doing something other than paying attention to the road. Of course this is not as attractive a selling point for robocars as enthusiastic claims of highly increased road capacity.

There are a couple things that I think could reduce human driving, perhaps quite quickly.

With robocars, there may be enough trusted evidence to prove negligence in causing an accient at a much higher rate than today, which could dramatically raise the cost of insurance for human drivers. Also, once robocars are better drivers than humans it's only a matter of time before the act of a human driving a car could be considered negligent.

Separately, drivers today will lose their license after a certain number of moving violations. We already have the technology to dramatically increase the enforcement of driving laws: it wouldn't take much to flag any speeding on highways, e.g. This could lead to many people not being able to keep their licenses even if they preferred to drive themselves. (not to mention how unpleasant it would be to drive if you had to make sure that you never went faster than the speed limit and followed all laws of the road to the letter).

In the USA. speeding is "illegal" but almost universal, and as a result, all attempts at systematic speed enforcement have been pushed back. You suggest that my robocar is going to narc out the car of the car next to me that's speeding? I think that's really, really unlikely. As are cameras everywhere doing it. Speed cameras and photo-radar mostly cause traffic jams when used on highways.

There is a rational approach, which is used in France and several other countries. The posted limit is 130kph (around 80mph) and it's the real limit too. In the USA the posted limit is 65 (sometimes 70 on rural interstates, and up to 80 in a few areas of Texas.) However, when the posted limit is 65 the de facto limit is around 75 to 80, but it's a soft limit, and depends on enforcement moods and budget. Which is of course silly.

Your other suggestion merits more thought. You are suggesting, as I see it, that people in collisions with robocars who are at fault will pay more because there will be clearer proof of their fault in the logs of the robocar's sensors. Those logs will indeed clearly show the story in most cases, but I am not sure what the percentage today of ambiguous accidents is, and so I am not sure if this is a big change.

While robocars logs offer certainty, I think most accidents today are judged fairly quickly by the attending police, and only a few have a big argument over fault. Certainly only a small fraction go to court. Insurance companies hate it if they go to court, and prefer no-fault if they can to avoid that, as go government insurance schemes. Robocar logged accidents should never go to court when it comes to determining fault under the vehicle code. (They may go to court for other reasons.)

For your claim to be true we need a serious fraction of accidents which are ambiguous now to flip the blame, and that does not seem likely to me.

For automatic driving law enforcement, I wasn't thinking of the existing speed cameras but rather if there were cameras at many locations that would simply log which car was where at some accurate time. Deducing speed would be trivial. Suppose every car entering and exiting 101 were logged, which could technically be done today. It would be trivial to determine average speed. This isn't 100%: for sure my average speed is <65mph during rush hour even if I speed for portions of my trip. But such a system could be put into place. Perhaps the main goal of such a system would be simply to reduce the amount of driving done by humans.

For proving negligence which would lead to large legal settlements I believe (admittedly without any real knowledge) I think it's not enough to establish fault. Knowing that one party caused a collision may not be enough but knowing indisputably how the other party was driving would likely increase the number of times negligence could be proven.

Also, standards of negligence change over time. If you are self-driving a car which is capable of robodriving and cause a collision, it may not be hard to prove negligence.

It's been tried. There are many rumours of it happening in the USA (apparently it has not) but it was tried in Scotland, and scrapped in 2007. Politically it does not fly. There are a few countries that still have them.

But again, the best system is to set the speed limit to the real speed limit, and enforce more seriously those who go above it, which becomes very few. It works.

Your proposal breaks the strong principle that you should not design technology to betray its owner. People hate it, and they will fight it and reject it, especially in the USA. (The systems put in place did not use people's cars, but cameras tracking licence plates.)

Add new comment