5 ms·

Since the article calls out Waymo engineers: I'm a software engineer at Waymo, speaking for myself. I agree with the premise of the article. Humans are remar

by jlebar 6y ago

Since the article calls out Waymo engineers: I'm a software engineer at Waymo, speaking for myself. I agree with the premise of the article. Humans are remarkably good at driving, and making a computer better will be hard. Where I may disagree with the author is, I think the project is both feasible and worth doing. It's worth doing because we drive a remarkable amount. Even at human safety level, more than a million people die every year in car crashes. As far as feasibility, Waymo is, today, running a fully autonomous, no-human-behind-the-wheel ride hailing service in Arizona. It's not a dead-end demo, it's there. I know, a Phoenix suburb is not the same as NYC or Mumbai. Trust me, I know -- probably better than you! Nonetheless, what we accomplished was impossible a few years ago. The history of AI is a history of moving the goalposts and then blowing past them. AlphaGo solves what was considered one of the hardest problems in AI, but now that I see how you did it, eh, it's not that impressive. GPT-3 passes the Turing Test and can almost pass a coding phone screen, but come on, nobody really thought that a machine that passed TT would be intelligent. It was just a few years ago that respectable people were saying we'd never solve protein folding. Now that we have, the goalposts have moved again. Don't worry, we'll blow past those, too. Except this time people will yawn. So give us some time. This too will happen slowly, and then all at once.

ljlolel 6y ago
I am the author, and I am a self-driving car engineer too! Waymo is an inspiring project, many friends work there, and it literally inspired me to work on self-driving cars. I wrote this to help people in the industry have the right context in the right way. As to Waymo's driverless testing, it's dangerous! See here what I wrote about why: https://jperla.medium.com/tesla-saves-lives-waymo-gambles-passenger-lives-3ee2d0e55f5f https://jperla.medium.com/tesla-saves-lives-waymo-gambles-pa...
pontus 6y ago
One of the issues that companies like Waymo will have to face is that it's not going to be enough to be better than a human driver, you actually have to be better than a human driver with lane assist, blind spot checkers, adaptive cruise control, etc. As self driving technology matures, so does driving assistance technology. The 100 Million miles that a human driver will have to drive on average to die in a crash will be pushed to a billion miles+ before L5 is achieved, effectively moving the goal post further. Ultimately self driving technology has to be better not than a human, but than human intervention. It has to be so good that it recognizes when a human incorrectly intervenes to override L4 decisions and then be able to override that human's overriding. Only then will autonomous vehicles overtake humans.
- ljlolel 6y ago
  Yes, and I write about that and Tesla's active safety features here https://jperla.medium.com/tesla-saves-lives-waymo-gambles-passenger-lives-3ee2d0e55f5f https://jperla.medium.com/tesla-saves-lives-waymo-gambles-pa...
- bbarnett 6y ago
  One real issue is that as automated systems, assist or full driving take over, human skill will decrease. Yes, there are people with poor coordination, spatial perception, the list goes on. Genetics makes a difference. Yet also, training makes a difference. So for those with normal or above normal genetic bounds for the above, training helps immensely. And driving, responding quickly to issues, is often about training. Hence all the driving ed programs now mandated in many localities, prior to obtaining a license. Still, it goes beyond that. One becomes more skillful, better at driving, as time progresses. At least, those capable of that. Meanwhile, we're moving humans further and further from direct learning experiences, and with each 'crutch', reducing training scenarios as well. Of course, we're on a path here. The goal is ultimately to move more towards safety. However I suspect that "things will get worse", before better, as drivers will be less trained on manual tasks. Edit: My main point is, we're going to get to a point of 'AI drives, you monitor/intervene" eventually, as the mainstay. Yet new drivers, will be less able to intervene. They won't have developed, or trained, via the same number of "close calls", and direct driving experiences. Eventually, asking drivers to "intervene", will be asking for the least skilled to take over, to monitor, to watch. And there are a lot of areas that are very high bar to handle with AI. Think on this ; all current driving tech is literally 'low hanging fruit'. It's been a lot of work, I grant that. Yet it's been a lot of work for just that low hanging fruit! For example, most AI driving is urban, in the southern US. How about: - In rural Canada - In the middle of a snow storm - With ice, and white-out conditions on the road - on rural dirt backroads, which are 20 or 30 km long, comprised of potholes and other such irregularities - with zero mapping - at -40C I have extremely strong doubts that AI can even remotely handle that well currently. I'd trust a 5 year old to handle this task far better. Now, what of drivers who never really drive, end up on this side road, and get 'stuck', with the AI unable to handle things? When all of the above is normal in the winter. For example, often schools do not close in rural Quebec, for such typical, and normal, conditions.
Mulpze15 6y ago
Why is the self driving ecosystem trying to solve every driving challenge, which is very hard, rather than focussing on highway driving, ideally reserved lanes, high speed and close cars. And progressively expand those areas "reserved to self driving card" over time. It feels like it would bring a lot of value while reducing complexity a lot. If this is a silly question, please let me know why.
- foreigner 6y ago
  I've wondered this too. Also start with commercial vehicles instead of consumers.
  - ljlolel 6y ago
    The hard problem with commercial vehicles is that they are even safer than the random consumers (teenagers, drunks, etc), so the bar is even higher (1 in a billion miles? 10 billion?) and has very quantifiable business metrics and insurance liabilities!!
    - mcguire 6y ago
      Is your proposal that we should take the most dangerous classes of drivers and make them more dangerous by giving them imperfect autonomous technology in order to improve the technology?
      - loopz 6y ago
        No need to put words and strawmen in mouth.
    - rhacker 6y ago
      I would like to see a computed stat, but I don't know, I see a lot of overturned semis often at these very specific places (elevation drop + turn) - always knocking them over. It's the wind or the speed the driver was going. Maybe iteration 1 of all this should be detectors that help drivers do better: 1. Unsafe driving speed alarms (given conditions) 2. Notification of areas that are historically unsafe 3. Baby in the back-seat! 4. Alcohol in the driver's seat detection (even if you don't have an existing DUI/setup) Any 1 of those implemented en-masse would likely reduce accidents/deaths world more than Full self driving AI (in its current form), and if that was truly the goal, I don't understand why these aren't the top little buggers to fix and release in 2021.
lhorie 6y ago
I'm legitimately curious about whether a physical hard limit exists for what AI can accomplish. A few decades ago, people often assumed Moore's law would continue indefinitely, but now there's some doubt as to whether it is physically possible to keep up the pace without some amount of marketing tricks. We don't believe AI would ever be able to solve some classes of "impossible" tasks, like breaking crypto. But at the same time, we say AI will just keep blowing past all hurdles in some slow singularity-like fashion. Those two ideas can't both be right. I'm curious where the lines converge at.
pron 6y ago
GPT-3 does not pass the Turing test. The Turing test is not about making a person think they're talking to another person while, in fact, they are talking to a machine. The reason is that such a result may depend on the expectations people have of machines' capabilities. Rather, the Turing test, in the "standard" interpretation (rather than Turing's original formulation) has an interrogator interviewing two subjects via text messages, knowing that one of them is a machine and the other a human, trying to distinguish which is which. The machine passes the test if it can trick interviewers sufficiently often.
- SAI_Peregrinus 6y ago
  This is a very important distinction. The Turing Test can be passed only by something that is indistinguishable from a conscious, thinking being.
- jlebar 6y ago
  > the Turing test, in the "standard" interpretation (rather than Turing's original formulation) Goalposts: moved! This is exactly my point.
  - pron 6y ago
    > Goalposts: moved! No, the original formulation is just a little unclear and some interpret it as an even more elaborate test, involving yet another stage. Most versions of the test include at least three participants [1]. Many people forget the details and remember just that it's about recognising a machine through text communication. [1]: https://en.wikipedia.org/wiki/Turing_test https://en.wikipedia.org/wiki/Turing_test
bmitc 6y ago
> It's worth doing because we drive a remarkable amount. Even at human safety level, more than a million people die every year in car crashes. I'm not sure. Why is this so? To save time? Sure, but I think the self-driving proponents that I have seen have not addressed the new problems that will be brought by self-driving cars. I don't think reducing crashes is anything but a marketing ploy because the real goal of self-driving is to make companies a lot of money. I think accidents will simply be transposed and not reduced. One problem is the initial rollout period where there is a non-homogeneous mix of human and automated drivers. This impedance mismatch will be problematic and likely lead to more accidents and congestion. In general, it's my belief (I don't have facts or research, just a hunch) that automated driving will actually increase congestion and traffic. We saw this with Uber and Lyft, where these things plus accidents were increased where these services were deployed. I've seen it first hand that the cars causing issues are likely to be Uber or Lyft. I will be very surprised if this is not the case with automated drivers. For example, consider the case of an automated car dropping someone off in a busy area, probably one of the most wanted use cases for automated car. This is because it could notionally drop the person off and then drive off and find parking. But that seems like an insanely hard problem because such a maneuver is actually difficult for humans and very tricky in busy areas. I truly don't understand the actual need for self-driving cars. There is a lot of hype, for sure, but I personally see that it will generate more problems. Basically the only problem that it will solve is reduction in human time driving because I doubt it will be actually safer such that it's a major difference from now. I feel our society would gain much more benefit in concentrating on better transportation and urban design: planes, trains, automobiles, bicycles, and walking.
- c22 6y ago
  I think your predictions are 100% accurate, but I also think as the fleet becomes more self-driving and as streets and other public infrastructure are redesigned to accommodate the new paradigms I believe these problems around traffic, congestion, and safety will be solved eventually by automated vehicles. That is to say I believe human-driving is at a local maxima and it will take accepting a negative dip if we want to progress on this front.
  - 6y ago
kurthr 6y ago
One of the key things that Self Driving Vehicles need to learn from humans is how to predict other humans/vehicles. Predicting which vehicle is going to drive unsafely, or even with excessive caution is not just responding to the position (speed and acceleration) of a vehicle, but how it acts relative to predictions (and how well it is predicting your actions), and fits into a type of driver model. And then knowing that all of that will change when a traffic cop is visible or not, the start of a rain, sunset, kids/pets acting up. That's why driving in different cities for an experienced driver can be so nerve wracking... it's hard to predict what all those vehicles will do! Meanwhile they are also predicting what you will do so you're dangerous as well (don't stop in traffic just because a pedestrian walks into the middle of a busy street in Delhi or Naples, you'll cause an accident).
dehrmann 6y ago
> I agree with the premise of the article. Humans are remarkably good at driving, and making a computer better will be hard. The economic incentives for autonomous driving are interesting, for but just saving lives, collision avoidance is where to focus. If you have tech as good as Waymo, integrating that into a car so it intervenes when it's highly confident the driver doesn't know what they're doing and highly confident it knows the right thing to do would be huge.

by jlebar 6y ago

ljlolel 6y ago

I am the author, and I am a self-driving car engineer too! Waymo is an inspiring project, many friends work there, and it literally inspired me to work on self-driving cars. I wrote this to help people in the industry have the right context in the right way. As to Waymo's driverless testing, it's dangerous! See here what I wrote about why: https://jperla.medium.com/tesla-saves-lives-waymo-gambles-passenger-lives-3ee2d0e55f5f https://jperla.medium.com/tesla-saves-lives-waymo-gambles-pa...

pontus 6y ago

One of the issues that companies like Waymo will have to face is that it's not going to be enough to be better than a human driver, you actually have to be better than a human driver with lane assist, blind spot checkers, adaptive cruise control, etc. As self driving technology matures, so does driving assistance technology. The 100 Million miles that a human driver will have to drive on average to die in a crash will be pushed to a billion miles+ before L5 is achieved, effectively moving the goal post further. Ultimately self driving technology has to be better not than a human, but than human intervention. It has to be so good that it recognizes when a human incorrectly intervenes to override L4 decisions and then be able to override that human's overriding. Only then will autonomous vehicles overtake humans.

ljlolel 6y ago
Yes, and I write about that and Tesla's active safety features here https://jperla.medium.com/tesla-saves-lives-waymo-gambles-passenger-lives-3ee2d0e55f5f https://jperla.medium.com/tesla-saves-lives-waymo-gambles-pa...
bbarnett 6y ago
One real issue is that as automated systems, assist or full driving take over, human skill will decrease. Yes, there are people with poor coordination, spatial perception, the list goes on. Genetics makes a difference. Yet also, training makes a difference. So for those with normal or above normal genetic bounds for the above, training helps immensely. And driving, responding quickly to issues, is often about training. Hence all the driving ed programs now mandated in many localities, prior to obtaining a license. Still, it goes beyond that. One becomes more skillful, better at driving, as time progresses. At least, those capable of that. Meanwhile, we're moving humans further and further from direct learning experiences, and with each 'crutch', reducing training scenarios as well. Of course, we're on a path here. The goal is ultimately to move more towards safety. However I suspect that "things will get worse", before better, as drivers will be less trained on manual tasks. Edit: My main point is, we're going to get to a point of 'AI drives, you monitor/intervene" eventually, as the mainstay. Yet new drivers, will be less able to intervene. They won't have developed, or trained, via the same number of "close calls", and direct driving experiences. Eventually, asking drivers to "intervene", will be asking for the least skilled to take over, to monitor, to watch. And there are a lot of areas that are very high bar to handle with AI. Think on this ; all current driving tech is literally 'low hanging fruit'. It's been a lot of work, I grant that. Yet it's been a lot of work for just that low hanging fruit! For example, most AI driving is urban, in the southern US. How about: - In rural Canada - In the middle of a snow storm - With ice, and white-out conditions on the road - on rural dirt backroads, which are 20 or 30 km long, comprised of potholes and other such irregularities - with zero mapping - at -40C I have extremely strong doubts that AI can even remotely handle that well currently. I'd trust a 5 year old to handle this task far better. Now, what of drivers who never really drive, end up on this side road, and get 'stuck', with the AI unable to handle things? When all of the above is normal in the winter. For example, often schools do not close in rural Quebec, for such typical, and normal, conditions.

Mulpze15 6y ago

Why is the self driving ecosystem trying to solve every driving challenge, which is very hard, rather than focussing on highway driving, ideally reserved lanes, high speed and close cars. And progressively expand those areas "reserved to self driving card" over time. It feels like it would bring a lot of value while reducing complexity a lot. If this is a silly question, please let me know why.

foreigner 6y ago
I've wondered this too. Also start with commercial vehicles instead of consumers.
- ljlolel 6y ago
  The hard problem with commercial vehicles is that they are even safer than the random consumers (teenagers, drunks, etc), so the bar is even higher (1 in a billion miles? 10 billion?) and has very quantifiable business metrics and insurance liabilities!!
  - mcguire 6y ago
    Is your proposal that we should take the most dangerous classes of drivers and make them more dangerous by giving them imperfect autonomous technology in order to improve the technology?
    - loopz 6y ago
      No need to put words and strawmen in mouth.
  - rhacker 6y ago
    I would like to see a computed stat, but I don't know, I see a lot of overturned semis often at these very specific places (elevation drop + turn) - always knocking them over. It's the wind or the speed the driver was going. Maybe iteration 1 of all this should be detectors that help drivers do better: 1. Unsafe driving speed alarms (given conditions) 2. Notification of areas that are historically unsafe 3. Baby in the back-seat! 4. Alcohol in the driver's seat detection (even if you don't have an existing DUI/setup) Any 1 of those implemented en-masse would likely reduce accidents/deaths world more than Full self driving AI (in its current form), and if that was truly the goal, I don't understand why these aren't the top little buggers to fix and release in 2021.

lhorie 6y ago

I'm legitimately curious about whether a physical hard limit exists for what AI can accomplish. A few decades ago, people often assumed Moore's law would continue indefinitely, but now there's some doubt as to whether it is physically possible to keep up the pace without some amount of marketing tricks. We don't believe AI would ever be able to solve some classes of "impossible" tasks, like breaking crypto. But at the same time, we say AI will just keep blowing past all hurdles in some slow singularity-like fashion. Those two ideas can't both be right. I'm curious where the lines converge at.

pron 6y ago

GPT-3 does not pass the Turing test. The Turing test is not about making a person think they're talking to another person while, in fact, they are talking to a machine. The reason is that such a result may depend on the expectations people have of machines' capabilities. Rather, the Turing test, in the "standard" interpretation (rather than Turing's original formulation) has an interrogator interviewing two subjects via text messages, knowing that one of them is a machine and the other a human, trying to distinguish which is which. The machine passes the test if it can trick interviewers sufficiently often.

SAI_Peregrinus 6y ago
This is a very important distinction. The Turing Test can be passed only by something that is indistinguishable from a conscious, thinking being.
jlebar 6y ago
> the Turing test, in the "standard" interpretation (rather than Turing's original formulation) Goalposts: moved! This is exactly my point.
- pron 6y ago
  > Goalposts: moved! No, the original formulation is just a little unclear and some interpret it as an even more elaborate test, involving yet another stage. Most versions of the test include at least three participants [1]. Many people forget the details and remember just that it's about recognising a machine through text communication. [1]: https://en.wikipedia.org/wiki/Turing_test https://en.wikipedia.org/wiki/Turing_test

bmitc 6y ago

> It's worth doing because we drive a remarkable amount. Even at human safety level, more than a million people die every year in car crashes. I'm not sure. Why is this so? To save time? Sure, but I think the self-driving proponents that I have seen have not addressed the new problems that will be brought by self-driving cars. I don't think reducing crashes is anything but a marketing ploy because the real goal of self-driving is to make companies a lot of money. I think accidents will simply be transposed and not reduced. One problem is the initial rollout period where there is a non-homogeneous mix of human and automated drivers. This impedance mismatch will be problematic and likely lead to more accidents and congestion. In general, it's my belief (I don't have facts or research, just a hunch) that automated driving will actually increase congestion and traffic. We saw this with Uber and Lyft, where these things plus accidents were increased where these services were deployed. I've seen it first hand that the cars causing issues are likely to be Uber or Lyft. I will be very surprised if this is not the case with automated drivers. For example, consider the case of an automated car dropping someone off in a busy area, probably one of the most wanted use cases for automated car. This is because it could notionally drop the person off and then drive off and find parking. But that seems like an insanely hard problem because such a maneuver is actually difficult for humans and very tricky in busy areas. I truly don't understand the actual need for self-driving cars. There is a lot of hype, for sure, but I personally see that it will generate more problems. Basically the only problem that it will solve is reduction in human time driving because I doubt it will be actually safer such that it's a major difference from now. I feel our society would gain much more benefit in concentrating on better transportation and urban design: planes, trains, automobiles, bicycles, and walking.

c22 6y ago
I think your predictions are 100% accurate, but I also think as the fleet becomes more self-driving and as streets and other public infrastructure are redesigned to accommodate the new paradigms I believe these problems around traffic, congestion, and safety will be solved eventually by automated vehicles. That is to say I believe human-driving is at a local maxima and it will take accepting a negative dip if we want to progress on this front.
- 6y ago

kurthr 6y ago

One of the key things that Self Driving Vehicles need to learn from humans is how to predict other humans/vehicles. Predicting which vehicle is going to drive unsafely, or even with excessive caution is not just responding to the position (speed and acceleration) of a vehicle, but how it acts relative to predictions (and how well it is predicting your actions), and fits into a type of driver model. And then knowing that all of that will change when a traffic cop is visible or not, the start of a rain, sunset, kids/pets acting up. That's why driving in different cities for an experienced driver can be so nerve wracking... it's hard to predict what all those vehicles will do! Meanwhile they are also predicting what you will do so you're dangerous as well (don't stop in traffic just because a pedestrian walks into the middle of a busy street in Delhi or Naples, you'll cause an accident).

dehrmann 6y ago

> I agree with the premise of the article. Humans are remarkably good at driving, and making a computer better will be hard. The economic incentives for autonomous driving are interesting, for but just saving lives, collision avoidance is where to focus. If you have tech as good as Waymo, integrating that into a car so it intervenes when it's highly confident the driver doesn't know what they're doing and highly confident it knows the right thing to do would be huge.