Hacker News Reader

Comment by csours | Hacker News Reader

I would bet 10k that the definition of Level 5 will change enough to make this a moot point, or that another new classification system will be introduced.

There are so many assumptions in this space that it's a fool's errand to try to list them.

I think the next paradigm shift will be in the cultural and regulatory perception, not in technology.

I have NO IDEA what that might mean, and I could be completely wrong! I accept that.

I think the technology has to demonstrate what it can do, and society has to come to terms with what the tech can do, and by doing those things we will painfully update our priors. Our existing prior beliefs will not be our new prior beliefs.

View on HN

Changing classifications aren't going to affect the bet, because both of these gents know what they're betting on.

If I could make a $10,000 bet with John Carmack, I'd go the way Atwood has. No way there will be fully self-driving cars by 2030. That's only 8 years away. Fully self-driving cars need one of two things:

A) All cars are self-driving, and are networked, and centrally controlled. No exceptions.

B) The AI is smart enough that we should be worried about the ethical implications.

If an AI can react to terrible human drivers as well as a good human driver can, then I'd invite that AI over for dinner.

A) also doesn't work unless all other road users (bikes, scooters, skateboards, pedestrians, etc.) are also banned from the roads. The other option is limiting the autonomous cars to tunnels or dedicated tracks, but then we're just reinventing trains again.

L4 is basically reinventing trains, with invisible tracks in the form of high-def maps. Even that is just so hard ;_;

Given how many people you see driving down the road staring at their phones, reinvention of trains is probably still a good thing for society.

Good point. On the other hand, if we could encourage people to use actual trains, that would be much better for the environment.

Why do you need A and B? Waymo is operating driverless taxis at Level 4 in Phoenix; what do you see as fundamentally missing (aside from expanding range) for Carmack to be able to win?

Those taxis can’t even reliably drive in the rain yet, hence Arizona for the staging grounds. So they’re a far cry from meeting the standard, unless rain is now considered a natural disaster.

From what I'm reading, L5 would need to handle the annual meeting of the summer tires club... Assuming that's a problem in NY or Chicago.

The claim is that you need A or B, not A and B (“Fully self-driving cars need one of two things:”)

The gap between level 4 and level 5 is huge.

Basically, if you can find videos online of Waymo giving up and requesting backup for a situation not severe enough for humans to do the same, then you are not really at level 5 yet. If you cannot rely on the car to handle driving though any circumstance you could reasonably expect your hired chauffeur to handle, well then you are not level 5 yet. Examples of situations where humans give up are hurricanes, and blizzards.

And there are plenty of videos of it stopping for situations where no human would ever be equally confused, plus a good handful of videos of the cars doing absolutely unacceptable things resulting in near misses.

Are you really that confident that by 2030 Waymo's roadside assistance program will be disbanded, or at least reduced to a killswitch, summoning a replacement car and a tow truck (because obviously the car must be broken if assistance was called)? Because if they need more than that, it sure isn't level 5 yet.

I’ve driven in one. Felt like I was driving with a 16 year old driver. I’d never drive in the rain with it. Through intersection with faulty traffic lights, and police directing traffic? Nope. It never got onto the freeway, and it avoided complicated driving/merging decisions by driving through neighborhoods.

I worry about the ethical implication of a self-driving car (a slave AI smarter than it's occupant) being forced to go to McDonalds.

You're assuming near perfect driving.

IMO, minimum requirement for level 5 driving is merely to match human driver accident rates.

> IMO, minimum requirement for level 5 driving is merely to match human driver accident rates.

Minimum requirement for nerds, maybe. Minimum requirements for ordinary folks will probably be more along the lines of "will be noticeably better than me on average" and "will not make any given mistake more frequently (or more severely) than I would".

i.e. I expect people a lot of people would want "strictly better than me on practically every axis" as their minimum requirement, not just "on par with me based on the average accident rate". People just aren't going to put up with seeing cars make dumb mistakes that they would never make, even if the accident rate ends up being good on average.

Not in the courts, no. If switching to self-driving reduces accident rates in 99% of situations, but in that last 1% someone gets killed, then there's a lawsuit waiting to happen.

Lawsuits happen with existing cars. Life goes on.

I thought an all self driving fleet today is statistically safer than humans

> If an AI can react to terrible human drivers

That's not sufficient - nowhere near sufficient. If an AI can react properly when kids are playing in a 5mph zone with cars parked on either side of the street, before their ball bounces into the street. If it can react appropriately to seeing a young kid clearly struggling to ride a bike. If it can understand that the tennisball bouncing across the road will be followed shortly by an enthusiastic dog... et cetera, et cetera.

I don't need an AI to compensate for shitty driving (taking a roundabout the wrong way). I need it to drive so as to protect non-vehicles that it encounters.

God it's going to be unbearable isn't it. There's going to be 5 more years of basically no progress on level 5, big tech will lobby to change the regulations so they can basically call anything self-driving and you'll have Tesla bros running around shouting "We did it! Fully Autonomous!" when really the cars are just doing a decent implementation of cruise control.

If we can take really good cruise control and change the environment sufficiently to the point we can claim "fully autonomous", then perhaps that's also a solution.

So, banning cars in cities, where they might encounter bikes, children, pets, etc.? Sounds perfect

I'm not sure changing labels and regulations will affect things much. Level 5 is basically tell the car anywhere you want to go and then have a snooze while it drives. Either that works or not, subject to some safety level, say less crashes than human.

For me, the shift in perception isn't going to happen before the technological leap.

The best driver aid systems I've seen are still just well-executed implementations of lane keep/change assist and radar cruise control- technologies that have been available in luxury cars for ~15 years. None of this stuff is actually new technology. It took decades to make existing driver aid technologies reliable and cheap enough to make them available to the average new car buyer. I have no reason to believe it will get easier from here.

I also have no reason to believe that the legal dilemmas behind autonomous driving features will go away. One of two things must happen- in fatal accidents caused by a car operating autonomously:

1. The driver (occupant?) will be liable. OR,

2. The manufacturer will be liable.

Option (1) is unprecedented, morally reprehensible, and would likely be overturned when some manufacturer eventually screws up on a Ford Pinto scale. Option (2) means that autonomous driving features will forever be hampered by the "keep your hands on the wheel!" warning bells that we already deal with today.

> 1. The driver (occupant?) will be liable. OR,

> 2. The manufacturer will be liable.

We already have big corporations that will risk huge liabilities for fatal vehicle accidents; they're called insurance companies.

So there's precedent for big corporations risking huge liabilities. Just instead of collecting an annual "insurance premium" they'll be collecting an annual "self-driving service subscription"

>... Ford Pinto scale...

Link for curious: https://philosophia.uncg.edu/phi361-matteson/module-1-why-do...

Thanks! Funny enough, the topic touches engineering, business, actuarial science and peoples life or death situation. All ending in courts inconconclusively. A huge part of human endeavors put together and nothing came out of them. Very representative of our species.

> "nothing came out of them."

I understand you to mean that there was not a dramatic result; I believe there were results - engineers and managers changed their goals to include more testing of components, IIHS and regulators also added testing. Automakers do respond to forced recalls. Reputational harm taken as seriously as direct fines and punishment.

Reply on news.ycombinator.com