Future of humans driving motor vehicles.

[Pete Owens]

Hi
AI isn't here yet, only the technology to implement AI, e.g. lane detection, auto parking, auto braking, etc
I don't think that AI will appear because people are struggling with the question of the moral decisions that AI will need to deal with, without necessarily recognising that people make instinctive rather than moral decisions in an emergency and live with the consequences
For example you are with your child in an AI car on auto-driver (or whatever it'll be called) and you're consequently not paying attention. A child runs out in front of your car close enough that you couldn't stop. The car has a choice; swerve and collide head-on with the oncoming truck and put the car's occupants at risk, or keep going and put the pedestrian at risk, and there isn't time for you to focus and decide. Which manufacturer would make that decision and brave the courts and media, when drivers can make the decision and accept the jeopardy?
Regards
tim-b

Oh dear - I didn't think think it would be long before the "trolley problem" was raised.
And the weakness of the case against computer control is highlighted by opponents raising frankly absurd moral judgement issues.

Your argument in a nutshell seems to be that because human control doesn't make moral decisions then you oppose computer control because it doesn't make moral decisions.

If you genuinely consider moral decisions an important feature of a motor vehicle control system then clearly that is a strong argument against allowing human control. As you point out, human decisions are based on instinct - and humans have a powerful instinct for self preservation above all other concerns. Given the choice between the lorry and the class of nursery children, human instinct is pretty much guaranteed to make the immoral decision every time. Even if you make the worst possible assumption that the computer is programmed like a human for self preservation above all else then that is no worse than the current situation.

However, the sort of hypothetical situation is vanishingly rare - and caused in the first place by a lack of care on the drivers part by creating a situation where a crash of some sort was inevitable in the first place. The computer will have judged the situation in advance, noticed that they were heading into a situation with no room to change direction so slowed to a speed that they could bring the car to a stop when they are so close to pedestrians that they could put themselves in your path with little warning. Of course this is how we should all drive anyway, unfortunately few humans do. A powerful argument for handing control to a computer.

[merseymouth]

Oh No!!!!!!!!!!! The human Gremlins will screw everything up, as they can' t stop tinkering.
All systems would have to go to a "No-Go" default to prevent tampering. Definitely needed as more than a few current already "Chip" their CPU's, unleashing more BHP.
Human nature to think that they can be the one to be in charge. MM

[tim-b]

Hi

Your argument in a nutshell seems to be that because human control doesn't make moral decisions then you oppose computer control because it doesn't make moral decisions.

No, I don't oppose it at all, and I don't think that I imply that
My point is that manufacturers (humans) will find it hard to programme AI to make moral judgements in situations where a human doesn't make moral judgements. Essentially manufacturers will be responsible for an intelligence that makes a decision to kill and I don't think that's acceptable to them at a corporate level
I doubt that humans will be capable of concentraing on a task that they aren't invested in, such as your vehicle driving by AI, and won't retain the necessary situational awareness to react in a timely manner to make decisions, and a car that drives slowly just to protect its manufacturer is neither use nor ornament and you might as well cycle; there's a thought
We have low-level AI now for specific functions, e.g. playing chess, and the AI learnt faster than scientists predicted, but driving is a higher level function, which isn't always apparent to judge by some overtaking It's the ability to learn that is necessary for drivers to develop skills, and for AI it would need to amend it's instructions to learn, which is a concern of people such as Stephen Hawking and Elon Musk (Tesla Cars). Who knows where a system that can learn and rewrite programming might go? I don't think that anyone wants to go down that route
Regards
tim-b

[tim-b]

Hi
Sorry, I got diverted...

Even if you make the worst possible assumption that the computer is programmed like a human for self preservation above all else then that is no worse than the current situation

If I'm interpreting this correctly, self-preservation of a computer over a life is surely the worst of all options. Asimov's three fictional laws make a lot of sense and have been adapted several times,
"First Law: A robot may not injure a human being or, through inaction, allow a human being to come to harm.
Second Law: A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
Third Law: A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws." Isaac Asimov 1942
Regards
tim-b

[fullupandslowingdown]

Very good points made. I suppose i was initially thinking particularly of a fully autonomous and independent AI driver, whereas if you have direct interaction between AIDs then problems like the mini roundabout which routinely stump humans, can be easily sorted. Whichever vehicle gets there first even by a millimetre, is given priority by the other AIDs, and because they're intelligent, they don't mind, don't even sweat it because they knew before they even they had optical sight of the others that they were approaching and so automatically adjusted their speeds so allowing all vehicles to flow through the junction smoothly. like how it is supposed to work with roundabouts but doesn't because of the rascals that insist on powering through like maniacs....

And the moral decision making? sure, if the system works as it should, then the accident should never come close to ever occurring. I feel that some of us have become a little too jaded by human morality. Every so often there is a story of heroic self sacrifice for the benefit of others. If all the AIDs are actively communicating with each other, then the only time when an accidental collision becomes inevitable is when a human does something unpredictable at the worst time. Even then, with steering and braking under computer control, the consequences would surely be less than with human drivers who would always be slower to react, and might hesitate to take an unusual course which does reduce the danger, but which only 2 machines thinking in coordination could calculate and act on in that minuscule window of opportunity.

This question about morality, I suspect that for the manufacturers at least it will be more about legal liability. i.e if an AID did decide to k̶i̶l̶l̶ sacrifice one child instead of another, who gets sued? the owner, the renter, the manufacturer, or the dead child's parents?

My main thoughts about AIDs becoming accomplished enough to replace humans, was the visual recognition bit. Current "driverless" cars use a combination of RADAR, LiDAR and optical cameras in an attempt to read the road ahead, and still cannot reliably see humans. Humans clearly aren't perfect either as evidenced by the legacy of SMIDSY et al. But in theory a human can see any human under any reasonable condition, and can see random other obstacles, road conditions etc. Maybe the one advantage an AID could have is seeing black ice.

There is one problem though which any driver of an electric vehicle will know too well. Standards. Most governments seem determined that the free market should drive standards, with the current (puns galore) chaos of having multiple different cables for different chargers. According to most drivers on the EV boards, they typically carry 3 different charging cables with them for anything other than a local trip. There is no physical compatibility between chargers, and they also have to register with multiple different apps etc to operate chargers owned by different companies. Some allow payment by phone, some need a specific charge card, some bank cards. Will individual governments let alone an international committee actually decide on one standard for communication between AIDs from different manufacturers? At the moment, I doubt it, but without a common standard I doubt any bodge to allow different standards to talk to each other will result in a 100% failsafe system.

As for human attitudes, well, the media tells us that there is a growing number of yungsters who actually aren't sold by the lure of fast driving. A combination of climate concern, cost of ownership and simply changing ideas about when and how they need to travel, seems to me that maybe in another 25 years time, diehard petrol heads will be in a minority like smokers today, and desperately clinging to their diminishing rights as drivers. Maybe. But as most of us know, in a consumer society, powerful business interests can wield considerable sway other the masses using the evils of advertising. If, and only if the youngsters can begin to see this insidious evil amongst us, can we hope for a brave new world.

[Pete Owens]

Hi

Your argument in a nutshell seems to be that because human control doesn't make moral decisions then you oppose computer control because it doesn't make moral decisions.

No, I don't oppose it at all, and I don't think that I imply that

OK so we agree that moral judgement is irrelevent to the task of controlling a vehicle - in which case why on earth bring it up?

The only relevant criterion is how good a computer is at controlling a vehicle compared to a human.

My point is that manufacturers (humans) will find it hard to programme AI to make moral judgements in situations where a human doesn't make moral judgements.

Or maybe not.

Perhaps if you maker it clear whether or not you personally think moral judgements are a important factor in vehicle control systems then we can focus the debate better.

Essentially manufacturers will be responsible for an intelligence that makes a decision to kill

Of course; it won't it will be programmed to avoid hitting things which will save lives - not kill people.
We are talking about vehicles - not weapons designs {FFE - family-friendly edit }.

and I don't think that's acceptable to them at a corporate level

Indeed - and since computers will be very much better at not killing people than are humans then failure to install a computer control system in a vehicle will result in killing people and be unacceptable.

I doubt that humans will be capable of concentraing on a task that they aren't invested in, such as your vehicle driving by AI, and won't retain the necessary situational awareness to react in a timely manner to make decisions,

Certainly the low ability of humans to concentrate is one of the reasons why computer control will be much better drivers than humans.

and a car that drives slowly just to protect its manufacturer is neither use nor ornament and you might as well cycle

Certainly I doubt there will be a market for cars that are programmed to drive too fast for the conditions so make a habit of crashing into things. Avoiding collisions is thus n the interests of the users of cars - the people they could potentially collide with and thus the manufacturers.

We have low-level AI now for specific functions, e.g. playing chess, and the AI learnt faster than scientists predicted, but driving is a higher level function, which isn't always apparent to judge by some overtaking

The chess analogy is very relevant to the debate over computer control because it was thought to be impossible for exactly the same reasons people use to argue against controlling cars. Because being good at chess was thought to involve "intelligence" the idea of a computer doing it was a challenge to the essence of our being. Of course as soon as computers became better at it than the best human grand-masters then we realise that all we are talking about is performing a vast number of very simple calculations to evaluate all the possible permutations of a particular move over the next 25 moves. The fact that computers do it better than humans means that you now consider it as "low-level".

Compared to chess driving is trivial. And we have already automated many of the tasks that human drivers find particularly difficult such as reverse parking, controlling the fuel mixture and feathering the brakes.

The control options remaining are very simple - steer - accelerate - brake.

It's the ability to learn that is necessary for drivers to develop skills,

And one of the reasons that humans are so bad at it that we each have to learn those skills individually - and we rarely experience and thus have the opportunity to learn from the unusual situations that lead to crashes. We drive the same road every day so learn how fast we can take a corner without loosing control - until one day their is an oil patch. We drive past parked cars every day and children do leap unexpectedly into our path - so we learn to drive too fast to cope when one day they do. We drive at high speed down unlit motorways at night for year after year and learn not to expect stationary objects in our path until one day there is.

Due to our limited powers of processing and observation we have to make lots of short-cuts and our assumptions in our decision making. We can only look at one thing at a time - and move our attention about 10 times a second. So our perception of a complete panoramic view is an illusion. We simply haven't evolved the ability to observe or process the amount of data necessary to move at high speed through a complex environment. These things are not intelectually difficult - it just involves a lot of sensors and processing - something that computers are good at.

The computer has time to consider all those what ifs (what if a child emerges from behind that car or that car etc) rather than do as a human and simply assume the road will be clear, and also when something unexpected does happen the computer will notice it much sooner (look at the stopping distances in the highway code and see how much is accounted for by reaction time)

and for AI it would need to amend it's instructions to learn, which is a concern of people such as Stephen Hawking and Elon Musk (Tesla Cars). Who knows where a system that can learn and rewrite programming might go? I don't think that anyone wants to go down that route
Regards
tim-b

I think you have been reading too much science fiction.

[tim-b]

Hi

The only relevant criterion is how good a computer is at controlling a vehicle compared to a human

I agree that computers can do many tasks better than humans, e.g. play chess and reverse park, but that isn't the only relevant criterion

Perhaps if you maker it clear whether or not you personally think moral judgements are a important factor in vehicle control systems then we can focus the debate better

I have made it clear, but manufacturers are interested in selling quality cars and maximising profit with unique selling points. Will they go down a route that leads to cars that all act and look the same, and have the manufactured ability to make a choice to kill as in my example? Take the climate change debate and the current push to ban sales of new diesel and petrol vehicles in 10-15 years time, and balance that against the example of manufacturers making large SUVs, and including the option of a petrol V8, a quad-turbo 3 litre diesel but without an electric/hybrid variant
Why would manufacturers invest in AI, a technology that potentially decimates their bottom line?
If you want to remove risk then infrastructure needs massive reform before AI hits the streets, which includes the way that we shop, etc. This could be more easily developed at sea with AI shipping, fewer, slower-moving hazards and radar that can sweep miles of open sea unimpeded. Prove the theory out there and its advantages and get the manufacturers on-board
In the same way as the H-debate this can and will go around in circles, so ahm oot
EDIT: to remove vehicle model specifics
Regards
tim-b

[kwackers]

The assumption is always that there'll be sudden turning point where there'll be a choice between human driven or computer driven cars.

Won't happen like that.

What will happen is automation will creep in and most folk won't notice it.
At some point cars will simply have the option to self drive, at the moment that's along motorways and such, in a few years it'll include A roads, a few years later B roads and eventually your entire journey.
There'll be no decisive turning point, simply a slow move towards automation and with it our trust will slowly be acquired.

It's impossible to get a 5 star NCAP rating these days without automatic brakes and even some cheap cars now have limited self driving ability (steer in lane, traffic following in jams, adaptive cruise control, cross traffic radar etc)

[willcee]

interesting comments.. however please bear in mind that this will likely only happen if it ever does on major highways and motorways. then take into account that to repair and bring those roads up to par again, install whatever magic needs to be in the tarmac and maintain them so will cost several billions , that on top of what the Westminster ingrates have thrown around to steady up the economics in this pestilence .. personally although i wouldn't ever be in favour, i feel that the general public have enough to contend with over the next few years than worry about self driving cars, and likely those boyos in power will eventually realise that..
And In Ulster where yesterday the minister said that it will take 1.2 billion to level out our wee bits of 3rd world tarmac back to par..go figure.. and our Gov Agency MOT sheds have shut up shop cause their hoists failed ,cracked at stress points having been used for 2/3 times their finite life, 'cause they wouldn't spend the money to replace them.. so MOT 's are unlikely to work for another year they just give you a note to extend your certificate...will...

[JohnW]

Yes, but the public outrage at the idea that people give up their *right* to drive their own cars....

.....................and their right to kill people? their right to inconvenience humans? their right to poison humans? their right to destroy the environment and eventually our habitat?.................never - it's their religion and it'll never, ever stop.

[[XAP]Bob]

AI surpassing human prediction has already happened. Most humans can’t predict anything not already on their bonnet.

[De Sisti]

Should humans be allowed to continue controlling motor vehicles on public roads once AI in driverless vehicles surpasses human controlled driving in safety and efficiency?

Can't ever imagine it happening in the US.

[Jdsk]

What will happen is automation will creep in and most folk won't notice it.

Overwhelmingly agree. But there'll be some special niches that don't follow that.

It's impossible to get a 5 star NCAP rating these days without automatic brakes and even some cheap cars now have limited self driving ability (steer in lane, traffic following in jams, adaptive cruise control, cross traffic radar etc)

See also the new EU standards, with descriptions of which technology by when.
https://www.europarl.europa.eu/cmsdata/155060/PPT%20General%20Safety%20Regulation.pdf

Jonathan

[JohnW]

AI surpassing human prediction has already happened. Most humans can’t predict anything not already on their bonnet.

..................shouldn't drive with anything on their bonnet - distracts their attention!

[JohnW]

..............What will happen is automation will creep in and most folk won't notice it...........radar etc)

..........a bit like 20mph speed restriction signs near schools...........