Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Speculation: Global Impact of Autonomous Vehicles

This site may earn commission on affiliate links.
Hello All - thought it would be fun to start a thread to share ideas & predictions on the impact of autonomous vehicles. Please share your thoughts on what the future might look like . . .

Here are some of my own thoughts on the topic:
  • Self-driving cars will become ubiquitous
  • Self-driving will lead to phasing out of human drivers (after >50% or so of cars on the road are self-driving, the human drivers will be the unsafe variable).
  • Self-driving will lead to fullly-autonomous vehicles, giving rise to Uber-like business models for passengers & delivery of goods
  • Uber-like availability of cars, with no drivers, will make on-demand transportation cheaper than owning your own cars & trucks
  • As on-demand autonomous vehicles gain share, car utilization rates will go up sharply (average in US today is ~5%). These cars could theoretically achieve 85%+ utilization
  • For each doubling of utilization rate, annual car sales could be cut in half (not including technology shifts and accelerated wear)
  • Car makers will innovate more quickly to survive, leading to even faster technology evolution (the end state comes sooner)
  • Eventually, only a handful of car makers remain
And if we think even further:
  • Automotive insurance companies will have to totally reinvent themselves
  • Garages will be re-purposed (maybe the storage business will decline as people opt to bring their stuff back home)
  • Mobile wifi-enabled offices for longer commutes will become commonplace
  • Living in the suburbs, farther from work, will be OK
  • Etc.
Fun, but also a bit ominous, to speculate on such things!
 
  • Informative
Reactions: voyager
I for one will not give up owning my car to ride around in some lifeless, nasty pod on wheels that I can't customize and takes to long to get to my driveway. By the time that autonomous tech gets that good, we will be able to have it so when the human is driving, the computer can shadow over the driver and keep things just as safe as a full autonomous car would.

I think a lot of people will feel the same way.
 
Also how do you get the package from the truck to the doorstep? I don't think car ownership will ever drop below 50%. Only place I thing it will drop a significant amount is in the middle of big cities. And it's already low there.
 
There are huge sections of the country (dare I say, >75%) where the idea of on-demand autonomous vehicles replacing private car ownership will never fly. It may work in cities or dense suburbs where public transportation is an option, but in my small NH town where I can't go anywhere without getting in the car? No way.

On Sunday I was getting ready to make a loaf of bread, and realized the butter had expired. I got in my car, drive 10 min to the grocery store in the next town, re-stocked, and headed home. Compare that to the "on-demand" model - I fire up the "bring me a car" app, wait, what, half an hour for a car to arrive, only to find that no one cleaned it out from Sat nights partying. The drunk in the passenger seat threw up on the floor. And I don't even want to know about the random bodily fluids on the back seat.

I'll pay for my own car just to maintain some sense of spontaneity and hygine in my life, thank you very much.
 
Thank you @n_coming and @tga for stepping in with the first comments. I agree my starting thoughts may lean more toward city situations. In any case, it would be great to hear others' ideas about what such a future looks like. Please share your thoughts on what the future of autonomous vehicles might look like in your areas.
 
Inthink in the longer term, 20+ years, all transport becomes autonomous and this is a huge upheaval in a lot of industries. There will still be a place for car ownership and human drivers will likely stay legal in rural areas. I think there will be a lot of pressure to take the last human drivers out of the mix in cities because they'll be holding back safety and efficiency. It's a really weird difference once the last humans are out. At that point you can get rid of stop lights and signs and traffic can just flow full speed in both directions through intersections. People could get where they're going much quicker and safer so would tend to vote human driving out. There would be a full mix of shared and personal transport at different rates. Bus and van service would be better and cheaper with autonomous vehicles and likely integrated in Mobilty apps. EG the cheapest option is shared ride vehicles that drop you off for pick up by other shared ride vehicles heading in the right way and coordinate it for you. You could send mobility app pickups and vans to get stuff like building materials without going with them. A lot more businesses would have app based ordering and delivery.

There will likely be a mix of different patterns of ownership and mobility services.

In the short term I'm really interested in who's first with a good integrated solution to autonomous vehicle/networked/Mobility app stack. I think Tesla is well positioned.

Looking at Uber Black/SUV/Luxury rates or rates for limos and black cars in general in major cities it seems like a fleet of Model X's on demand with a Tesla Mobility app could be enormously profitable.

Putting 50,000 Model X's to work in Mobility Fleets could generate $200k to $500k each in profit per year or $4B to $10B. An order of magnitude higher returns than selling the same vehicles.

Tesla or any other OEM in place first could take over markets from Uber pretty literally over night in the same pattern Uber followed, one market at a time with full saturation on day one.

Uber can see this coming of course so is very motivated to develop it's own stack with partners, it looks like Toyota is the manufacturer. They have no battery electric platform and no OTA update potential in any existing platform (which matters a lot to developing and refining a system).
 
There are huge sections of the country (dare I say, >75%) where the idea of on-demand autonomous vehicles replacing private car ownership will never fly. It may work in cities or dense suburbs where public transportation is an option, but in my small NH town where I can't go anywhere without getting in the car? No way.

On Sunday I was getting ready to make a loaf of bread, and realized the butter had expired. I got in my car, drive 10 min to the grocery store in the next town, re-stocked, and headed home. Compare that to the "on-demand" model - I fire up the "bring me a car" app, wait, what, half an hour for a car to arrive, only to find that no one cleaned it out from Sat nights partying. The drunk in the passenger seat threw up on the floor. And I don't even want to know about the random bodily fluids on the back seat.

I'll pay for my own car just to maintain some sense of spontaneity and hygine in my life, thank you very much.

A lot of people will prefer to own a vehicle, especially in less populated areas. There might be Vehicle sharing apps allowing neighbors to coordinate borrowing vehicles. A vehicle could be sent on it's own to perform many tasks you have to do now. The grocery store would have a drive through for autonomous vehicle pickups. You could order groceries or whatever on your mobile device and send your vehicle to pick it up. Click the box for returns/exchanges put the butter in the trunk and send your car to do that chore for you.

However people might not need to own as many vehicles. You don't necessarily need one for everyone in the household if it can drop dad off for work then come back and take care of other trips instead of sitting in a parking lot all day.
 
The two biggest hurdles to vehicle autonomy are cost and our legal tort system.

Our legal system is not designed to permit "acceptable failure rates", it is mapped out for deep pockets instead. Tack on additional legal costs, additional safety armor, and the autonomy system itself, and expect a truly viable and profitable autonomous car to operate at a rate several times higher than our existing cost per mile.
 
For Mobility app services I suspect the Terms of Service will include agreeing to extra fees if a passenger messes up the interior as in the scenario. I'm guessing a fleet would contract with car washes around town for at least daily inside and outside service. In non peak times the vehicles would just show up and present themselves for cleaning. The fleet could bargain for good rates because they could fill down time at the car wash.
 
The two biggest hurdles to vehicle autonomy are cost and our legal tort system.

Our legal system is not designed to permit "acceptable failure rates", it is mapped out for deep pockets instead. Tack on additional legal costs, additional safety armor, and the autonomy system itself, and expect a truly viable and profitable autonomous car to operate at a rate several times higher than our existing cost per mile.

I would have thought that too a few years ago but have come to think it's likely less of a barrier.

The OEMs that operate the fleets will have deep pockets and experienced legal teams and ALL the data. Any accident will be way more documented than any accident with human drivers. There will be some cases when autonomous vehicles are at fault and the OEMs will settle generously. Otherwise they will beat frivolous claims into submission. Keep in mind these vehicles will have cameras in every direction, ultrasonic sensors, radar and all that data will fit with a tremendous depth of understanding of how the vehicle performs. Most legal teams won't be equipt to deal with this. The ones who can will know they're usually better off with a settlement. My guess is it will come to be seen as not worth the effort by most lawyers. There's an asymmetry here. You're sueing a big very profitable company over it's core competency that it's extremely well prepared to defend.

A related model for this is Uber. If you described their business model 10 years ago anyone who know how things worked would have laughed at it. It's a liability nightmare and walking into a major battle with powerful entrenched interests in every market. It didn't turn out like that. Uber is prepared for the fights. An OEM who does this would be too.

Tesla's example is pretty critical. Nobody else is even in the game right now collecting data. They're pulling in data every 10 hours for miles equal to Google's whole history of driving autonomous test vehicles. When the time comes to switch on Mobility App fleets they'll be very confident in the data ...likely enough to feel that what seems very bold on liability will be a very well understood and manageable risk.

Just switching on autopilot was a bold move, but they understand real risks better every day now.
 
Last edited:
Interesting discussion. Much to speculate about. The IEEE says it will take at least until 2040 before we see reliable autonomous cars. Cars driven by their owners will co-exist for decades. Question is whether people-driven vehicles will interfere with the whole process of traffic becoming autonomous, will it lead to extra problems. Will we see outlawing people-driven vehicles any time in the future? I don't think so. Too much resistance to change. To start with, a car is an undeniable source of pleasure... not just something utilitarian.

The opportunities are limitless, we all agree. Detroit uses self-driving features to extend shelf life of the car as we still know them today. Silicon Valley sets out to 'grab it all' - personal mobility catered for in all of its aspects. Including public transport-like services. Strangely enough, there is still too little effort to come up with a new transportation mode that is to take on all of the new roles that van be expected. Google and Apple (search for the i-Car) are trying to.

Copy%2Bof%2BFusing%2Bprivate%2B%2526%2Bpublic%2Btransport%2B2.jpg
 
Last edited:
I always love these threads because there are always so many people that object to autonomy.

30 mins to get the car? Crazy. 90% of people live in appropriate density to get a car quickly. Doesn't mean that poster did but still. I live in a moderately dense area (3000/sqmi). Uber is always less than 10 min and I anticipate the shared/autonomy model as at least 1/2 the uber time - so typical would be 4 min.

Now if I needed butter - I wouldn't leave the house. I might have to go outside and get from the delivery truck when my phone texted me. I anticipate that taking 10 minutes.

Driving is fun - 5% of the time. I've owned RX-7, 300TT, 5 series, Model S - all fun cars to drive. But the 95% makes it not worth it. Please let me get work done while in the car.

The problem with letting humans drive is traffic. I think this is under appreciated. When humans cause traffic as compared to the robot with millisec reaction time, they should be outlawed. Now perhaps there becomes a rule that allows humans to drive when they aren't holding me up - that would be fine.

I live in a modest traffic area. But put everyone in computer controlled EVs that accelerate to 40 mph 2 secs after a light turns green and traffic is way way down. Put following distances at 2 car lengths with zero variability. Zero person reacting slowly since they are on a phone. Zero accidents slowing traffic. Zero rubber necking even for the rare accident (tree falling across road etc).

Then, reduced energy use. Reduced fatality, reduced health care cost from car accidents. It is truly transformational. And most people don't get it.

Car utilization will never be 85% though. We all go to work/school at about the same time. That is still an issue. But utilization if now 5%, could get to 10%. Multiple car households could certainly reduce a bit. But the wealthy will still choose to have their own car for the cleanliness/privacy of it all. But you also have more potential buyers with autonomy - my 6 yo for one. A blind person. An 80 yo who wasn't comfortable driving anymore.

The car culture is slowly dying anyway. But we will still need to get to work.
 
  • Like
Reactions: bjelkeman
Thanks everyone for continuing the discussion. A couple of additional thoughts . . .

- If autonomous vehicles work on streets, why not drones to deliver that butter you forgot, directly from your local grocery store?
- Agree that average car utilization probably won't go to 85%, but expect the Uber-like vehicle could get that high
- The one car family may become the norm again . . . family car takes me to work, then it goes home for spouse & kids to use, then it comes pick me up at the end of the day. This I believe is realistically doable within the next 10 to 15 years.
- If we continue to own our own cars, there is also the idea that a vehicle could become an integral part of our houses, such as an office or a media center or whatever. Imagine it "docking" to the house in a functional way that provides further use when integrated to the home.

"The future ain't what it used to be!"
 
Lets see... some random thoughts on various points here:

One thing not covered here that I would predict is that if we went down this road, cars will become vastly more expensive... along the lines of 100k-200k in today's money for starter cars. Much of this could be in fees to cover the world wide traffic control system needed, but also because the cars themselves will need to be much more complex and much better tested.

However, this would not directly impact the majority of users since they would simply pay for the part of the time they needed it either in per hour/mile/kilometer fees, or in some kind of car sharing arrangements.

Also how do you get the package from the truck to the doorstep? I don't think car ownership will ever drop below 50%. Only place I thing it will drop a significant amount is in the middle of big cities. And it's already low there.

In most areas the post office is requiring residents to install these combined mail boxes to make their job easier. I could see a time where they were automated and an autonomous vehicle pulled up, dropped off a package in box 123, and texted you that the package is there, the code for the door is "abcd".

I can also easily see it going below 50%... keep in mind that according to the 2010 US census 80.7% of people lived in urban areas. (not saying it would, but not willing to rule it out either)

On Sunday I was getting ready to make a loaf of bread, and realized the butter had expired. I got in my car, drive 10 min to the grocery store in the next town, re-stocked, and headed home.

Consider this: https://www.amazon.com/b?node=8037720011, far fetched, but certainly possible. Also, with a decent network of autonomous vehicles, you could have one that stopped by the autonomous store and dropped that butter off at your place, probably quicker than if you went and got it. (for most people at least)

Remember where we were 20 years ago in terms of getting things you needed... overnight shipping was reserved for critical packages and cost a TON of money... now in many areas of the country I can get Amazon to deliver things in an hour or less...second day air shipping is commonly free! and 20 years from now, who knows where we will be!

Compare that to the "on-demand" model - I fire up the "bring me a car" app, wait, what, half an hour for a car to arrive, only to find that no one cleaned it out from Sat nights partying. The drunk in the passenger seat threw up on the floor. And I don't even want to know about the random bodily fluids on the back seat.

True, that's a challenge...but I suspect we would end up with tiers of service... there will be the mass transport option that has plastic seats and can wash out the interior with automatic hoses, all the way up to the executive transport ones that have plush leather interiors, full internet access, and conference rooms. And some people would simply still have their own vehicles to use if they could afford it.

Also, unless you lived WAY out there, say, on an island with only 20 people, or in the middle of Montana with no one for miles... you would likely have some kind of agreement with your neighbors for the car sharing service, so the car might only be a few doors down, and there would be 4 or 5 of them to make sure they were available. If you were one of the 1% (as are many Tesla owners), you probably would have your own car, but that would probably not be the norm.

In addition, I suspect that we should keep in mind that over time people's ways of life will change (kind of like going from a gas vehicle to an electric one)... people will not expect to just be able to jump up and run out in an instant. (like we have to plan better to make sure we are always charged up)

The two biggest hurdles to vehicle autonomy are cost and our legal tort system.

Our legal system is not designed to permit "acceptable failure rates", it is mapped out for deep pockets instead. Tack on additional legal costs, additional safety armor, and the autonomy system itself, and expect a truly viable and profitable autonomous car to operate at a rate several times higher than our existing cost per mile.

Also true... but remember at least in the US, for good or bad, the legal system will often change to handle challenges like this, with limitations being implemented to help shield companies (or people) from liability. When there is enough money on the line, the wheels of justice can move much quicker (though not always in the direction we want them to).

The IEEE says it will take at least until 2040 before we see reliable autonomous cars.

The IEEE is on crack! *(I suspect I am taking their statement out of context though) we are really close now, and I suspect within the next 3-5 years we will be able to see reliable cars that will handle enough situations that they can be considered usable in lots of places. That doesn't mean we will have the litigation and regulatory approvals to allow it to happen, nor that it could (or should) happen everywhere, but the technology is close enough today that I can see it, and I don't see any real problems left to overcome that we don't have answers for... yes, it will still take time, lots of testing and learning, and perhaps more sensors and processing power than we throw at it today, but nothing that I have heard anyone complain about is "technically" impossible anymore.

Cars driven by their owners will co-exist for decades. Question is whether people-driven vehicles will interfere with the whole process of traffic becoming autonomous, will it lead to extra problems. Will we see outlawing people-driven vehicles any time in the future? I don't think so. Too much resistance to change. To start with, a car is an undeniable source of pleasure... not just something utilitarian.

Totally agree, owner driven cars will remain for decades, but I suspect, not that many decades... If we are talking about 20+ years from now, I could see owner driven cars banned from interstates and cities, and if you want to drive, you will have to turn on your autopilot to take you out of the city to RT 101, or US 9, or whatever other favorite driving road you have, away from the commercial routes and paths. (on the other hand, it might make for better driving for the enthusiasts since there would probably be less people out there.)

When making guesses at the future, we need to keep in mind that over time it's not just the technology that changes, social attitudes and society changes as well, and not always the way we would like to see!

There... all done making guesses and judgments that I don't really know enough about to make!
 
  • Like
Reactions: Roger_wilco
In initial phases , some roads will be classified as autonomous roads (there will be some gadgets to support autonomous driving). Using autonomous roads will cost you something, to prevent unfair use .
Cargo companies will make great use of autonomous cars. they will start serving not only for regular cargos but even for your daily or weekly shoppings. Large markets will install robotic collection systems for goods to complete your shopping list in shortest time and load to autonomous delivery vehicles. This way, cars will serve to transport you if you really have to be there. In other cases , things will come to you.
Manual driving will always remain but driving licenses will be hard to obtain items for teen individuals of autonomous generation. They will question about why/when it will be needed, even a flat tyre can be changed by autonomous and automatic tyre repair vehicles.
And finally, obesity will become primary cause of human deaths.
 
An actively supervised autonomous vehicle is safer than an unsupervised autonomous one.
An actively supervised autonomous vehicle is safer than an unsupervised autonomous one.
An actively supervised autonomous vehicle is safer than an unsupervised autonomous one.


Once a human operated vehicle gains robot emergency braking, robot lane assist, V2V communication and GPS, pretty much all the safety benefits are captured, the benefits of Autonomous cease to be safety, but really are convenience instead.

But it will be a very desirable convenience feature, so autonomous capable, but privately owned vehicles will be the pinnacle.

Vehicles are like couches and toilets and kitchens are private places. Autonomous vehicles will enable society to own more vehicles, not less. Its part of Jevrons paradox, the more useful and efficient a resource is, the more the resource is used by society.
 
  • Like
Reactions: Roger_wilco
Vehicles are like couches and toilets and kitchens are private places. Autonomous vehicles will enable society to own more vehicles, not less. Its part of Jevrons paradox, the more useful and efficient a resource is, the more the resource is used by society.

The limitation on ownership is cost. Autonomy increases the cost of individual vehicles, but offers improved and cheaper alternatives to ownership. Unless autonomous systems end up being very cheap, they should be expected to reduce private ownership, with the most obvious effect being reduction of the number of multi-vehicle households.
 
Car utilization will never be 85% though. We all go to work/school at about the same time. That is still an issue. But utilization if now 5%, could get to 10%. Multiple car households could certainly reduce a bit. But the wealthy will still choose to have their own car for the cleanliness/privacy of it all. But you also have more potential buyers with autonomy - my 6 yo for one. A blind person. An 80 yo who wasn't comfortable driving anymore.

The car culture is slowly dying anyway. But we will still need to get to work.

Full autonomy can change a lot of things, and commuting is definitely one of those things. It can merge taxis, road mass transportation, rental and car-sharing into a single system. With no driver to hire, no car to pick-up and no car to return, what's the difference between a taxi and rental? With no drivers to pay, why have the inconvience of restriction of large buses? If people are using apps to get from and to places, why not do some route matching and picking up and dropping off along on the route to share costs?

While there's a lot of 8-5, there are also a lot of shift workers. For every instance where there are n shifts per day, you can multiply utilization by n and if those shifts don't match the 8-5, divide by n+1.

Have a choice: pay full price and ride on your own, or add 5 to 10 minutes to your commute, share the vehicle and save 40%. Add another 5 to 10 minutes and save another 20%. Or walk to a ride-share stop and save 80%.

Buses become faster, more efficient and more convenient: divide the bus into 5 minibuses. If the minibus is full it only needs to stop when someone wants to get off. So buses can become smaller, and there can be more express commuter services. And with more buses, they can run more frequently to areas nearer peoples' homes.

There possibilities are exciting, and the potential synergies with electrification are also exciting.
 
  • Like
Reactions: Model 3