Self-driving technology is going to change more than cars

I'm not holding my breath. I grew up hearing people talk about how we would have Mars colonies by now. Imagine my surprise a couple of decades later when the shuttle program ended and the U.S. lost the ability to send people into space for a time.

The way I see it, the clock is ticking. If self-driving cars don't reach a certain level of capability soon, I suspect they simply won't; the technology will have plateaued at a point where it will not be viable in the foreseeable future. It'll be like supersonic flight: full of promise in the 1960s and 1970s, but mothballed for decades after that waiting for some engineering breakthrough to make it commercially viable.

Trains. Such concepts have been around for a long while in rail travel. Implementations never go smoothly. The reality of automating fast moving boxes full of living people are stark. Even where trains are driverless, with extensive signalling and redundant navigation systems, they don't get closer/faster than human operated trains. As systems grow more complex, to the point that failures are a regular occurrence, the dream of hyper-efficient automated travel disappears.

Insurance.

If there is a significant reduction of cost from an insurance perspective, the cost to insure a human-driven vehicle will increase substantially to the point that it will become a burden to own a vehicle that isn't driven by software.

That is, unless the government feels the need to intervene here. But, business seems to have the ear of government a lot more than the citizens do now days.

The only tipping point we're nearing is when people finally realize that 80% of a self-driving car may as well be zero percent, that the last 20% is way more complicated than anyone wants to admit, and that even to they extent they ever "work", flooding our streets with robot cars is not the ticket to utopia.

I am skeptical that AIs capable of piloting fully driverless cars are coming in the next few years. In the longer term, I'm more optimistic. There are definitely some fundamental breakthroughs which are needed (with regards to causal reasoning etc.) before "full autonomy" can happen -- but a lot of money and creativity is being thrown at these problems, and although none of us will know how hard the Hard problem is until after it's been solved, my hunch is that it will yield within this generation.

But I think that framing this as an AI problem is not really correct in the first place.

Currently car accidents kill about 1.3 million people per year. Given current driving standards, a lot of these fatalities are "inevitable". For example: many real-world car-based trolley problems involve driving around a blind curve too fast to react to what's on the other side. You suddenly encounter an array of obstacles: which one do you choose to hit? Or do you (in some cases) minimise global harm by driving yourself off the road? Faced with these kind of choices, people say "oh, that's easy -- you can instruct autonomous cars to not drive around blind curves faster than they can react". But in that case, the autonomous car just goes from being the thing that does the hitting to the thing that gets hit (by a human). Either way, people gonna die -- not due to a specific fault in how individual vehicles are controlled, but due to collective flaws in the entire premise of automotive infrastructure.

So the problem is that no matter how good the AIs get, as long as they have to interact with humans in any way, they're still going to kill a fair number of people. I sympathise quite a lot with Musk's utilitarian point of view: if AIs are merely better humans, then it shouldn't matter that they still kill a lot of people; the fact that they kill meaningfully fewer people ought to be good enough to prefer them. If this is the basis for fostering a "climate of acceptance" then I don't think it would be a bad thing at all.

But I don't expect social or legal systems to adopt a pragmatic utilitarian ethos anytime soon!

One barrier it that even apart from the sensational aspect of autonomous-vehicle accidents, it's possible to do so much critiquing of them. When a human driver encounters a real-world trolley problem, they generally freeze up, overcorrect, or do something else that doesn't involve much careful calculation. So shit happens, some poor SOB is liable for it, and there's no black-box to audit.

In contrast, when an autonomous vehicle kills someone, there will be a cool, calculated, auditable trail of decision-making which led to that outcome. The impulse to second-guess the AV's reasoning -- by regulators, lawyers, politicians, and competitors -- will be irresistible. To the extent that this fosters actual safety improvements, it's certainly a good thing. But it can be really hard to make even honest critiques of these things, because any suggested change needs to be tested against a near-infinite number of scenarios -- and in any case, not all of the critiques will be honest. This will be a huge barrier to adoption.

Another barrier is that people's attitudes towards AVs can change how safe they are. Tesla has real data showing that Autopilot makes driving significantly safer. This data isn't wrong. The problem is that this was from a time when Autopilot was being used by people who were relatively uncomfortable with it. This meant that it was being used correctly -- as a second pair of eyes, augmenting those of the driver. That's fine: it's analogous to an aircraft Autopilot when used like that. But the more comfortable people become with Autopilot -- to the point where they start taking naps or climbing into the back seat -- the less safe it becomes. This is the bane of Level 2 and 3 automation: a feedback loop where increasing AV safety/reliability leads to decreasing human attentiveness, leading (perhaps) to a paradoxical overall decrease in safety and reliability.

Even Level 4 and 5 automation isn't immune from this kind of feedback loop. It's just externalised: drivers in Mountain View learned that they could drive more aggressively around the Google AVs, which would always give way to avoid a collision. So safer AI driving has led to more dangerous human driving.

So my contention is that while the the AVs may become "good enough" anytime between, say, now and 20 years from now -- the above sort of problems will be persistent barriers to adoption. These problems can be boiled down to a single word: humans. As long as AVs share a (high-speed) domain with humans, there will be a lot of fatalities, and the AVs will take the blame for this (since humans aren't black-boxed).

Nonetheless, I think we will see AVs become very prominent. Here's how:

1. Initially, small networks of low-speed (~12mph) Level-4 AVs operating in mixed environments, generally restricted to campus environments, pedestrianised town centres, etc. At that speed, it's possible to operate safely around humans even with reasonably stupid AIs. Think Easymile, 2getthere, and others.

2. These networks will become joined-up by fully-segregated higher-speed AV-only right-of-ways, either on existing motorways or in new types of infrastructure (think the Boring Company).

3. As these AVs take a greater mode-share, cities will incrementally convert roads into either mixed low-speed or exclusive high-speed. Development patterns will adapt accordingly. It will be a slow process, but after (say) 40-50 years, the cities will be more or less fully autonomous (with most of the streets being low-speed and heavily shared with pedestrians and bicyclists).

Note that this scenario is largely insensitive to AI advances, because the real problem that needs to be solved is at the point of human interface.

1: https://news.ycombinator.com/item?id=17170739

I wonder which of these will come true. Some of them, in part, I'm fairly confident of.

I think the police will switch from using driving as an excuse, to using online activity as an excuse. "This guy has been googling the same stuff that ISIS does, better bring him in!" And then they send a signal to whatever self-driving car you're in and tell it to lock the doors and drive you straight to the station for questioning.

Total vehicle miles will probably increase, but fleet vehicles tend to require less maintenance because they're more regularly maintained on average (I'm told.) Maintenance will probably also become more of a fleet/B2B thing than B2C "individual car problem" thing. Partnerships will be formed or maintenance will be brought in-house, and margins will go down.

I guess it's too much to expect Americans to act like the Japanese when it comes to taking care of public spaces.

It doesn't seem that car2go-style cars get vandalized a lot; next-user reporting controls most of what goes above a level of vandalism (smoking users and some trash in the car is normally not reported).

Rage against uber by taxis has been more rampant (imho); so if we add bus drivers, lorry drivers, taxi and Uber drivers, etc. there might be a lot of damages on the short term. But once these displaced workers find another job that vandalism will disappear, so it will be transient during adoption times only.

Would Detroit Michigan's snowy winters count? Google, Ford, and Chevy already do routine winter testing there : https://www.bloomberg.com/news/articles/2017-10-26/alphabet-...

Not only is there road-rage potential, think about the municipal revenue from traffic enforcement. The public purse will remain thirsty for the funds, but autonomous cars will be very difficult to fine (either because they don't break the law or because every ticket gets disputed with hard data from companies that can afford to defend themselves). So what will happen? People driving older cars will start to become "over fished". As time goes on, it will only be poor people getting tickets (even more so than today) and therefore disproportionately minorities, and the social unrest between those people and an antagonistic police force will worsen.

The way laws are made is an interesting and complicated topic. Most people have an innate understanding as to what is "correct", when when the laws don't match up then there are problems. They keep the speed limits a little below that number because then the police can pull over cars they don't like at will, and because of the ticket revenue.

Honestly, I think once autonomous vehicles pass the 50% mark, they will have to increase the speed limits just to keep everyone safe.

People want to go what is a safe and reasonable speed for the conditions (55mph is safe but not reasonable for an interstate highway in most non-traffic jam conditions), the driver-less car operators want to be able to get people and things from A to B just as quickly and the highway engineers know that having two groups of traffic following different rules on the same infrastructure is a recipe for disaster.

The law will be what change because changing the law doesn't require millions of dollars like creating separate infrastructure does.

As a bonus the people who follow the law without considering their surroundings (e.g. people going the speed limit in the left lane) will be less annoying to the majority of the population when the letter of the law becomes aligned to how the majority of the population acts.

I suspect this is the only way we are going to completely solve the technical problems. Something like this:

https://www.youtube.com/watch?v=Byk8LcPovOQ

Same goes for groceries, I suppose. A co-op of farms operate an automated warehouse together whose responsibility is to consolidate the various fruits and vegetables and distribute via autonomous vehicles to homes. It gives a competitive advantage to regional farms so they can actually still do things at a reasonable economy of scale.

Just my two cents. Also, yes, video games all day. :)

https://www.youtube.com/watch?v=iogFXDWqDak

All the supermarkets here have been doing home deliveries for years.

1) Fridge finds several items low and proceeds to order groceries.

2) Website schedules car to be driven to store.

3) Store Robot packs groceries and finds car in car-park.

4) Negotiation between Car and Robot occurs, groceries placed in trunk.

5) Car drives off home.

6) Car notifies domestic robot that groceries have arrived.

7) Groceries put into fridge by domestic robot.

8) Lazy human requests that domestic robot gets them a can of beer. Unaware that the whole process occurred.

I've long wanted this, travel for 10 hours at night while I'm asleep, check out a historical museum in middle of nowhere Kansas, drive an hour, have lunch, and check out some county park (again in the middle of nowhere), then have supper, drive a bit and watch the sunset over some historical marker and the process repeats until I'm at Grandma's house. Much better than the current plan of watching cotton fields go by.