I feel that this sentiment could be applied to almost anything and isn&#x27;t really meaningful. Almost every accident ever could be attributed to negligence somewhere.

In NYC the urbanists have a trend of saying &quot;crash&quot; instead of &quot;accident&quot; when referring to traffic collisions. They feel that &quot;accident&quot; gives the driver too much credit in situations where they were completely at fault for not looking, going too fast, running a light, etc.

Agreed; more generally, there&#x27;s a movement to avoid calling these incidents &quot;crashes&quot; instead of &quot;accidents&quot;, since it&#x27;s arguable that crashes are avoidable: <a href="https:&#x2F;&#x2F;www.crashnotaccident.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.crashnotaccident.com&#x2F;</a>

I agree completely. Here, there&#x27;s perhaps negligence of the system, and the driver that was supposed to be overriding it in situations like this.

This is a dumb saying that ignores the fact that humans are fallible beings who make mistakes.Sure, you might shoot yourself because you&#x27;re just careless with your gun. Or you might be also thinking of an argument you had with your spouse, or you might be tired when you&#x27;re handling your everyday carry on your way to work, or you might get startled, or a million and one other things that might prevent you from singly focusing your attention the gun you&#x27;re handling.Machines exist to serve us, not the other way around. If a product doesn&#x27;t take human habits and flaws into account, it&#x27;s a bad product, because over a large enough sample size someone will make a mistake.

There&#x27;s an old saying around gun owners: &quot;There&#x27;s no such thing as a firearms accident. Only negligence.&quot;I think the same thing applies to autonomous things.

They&#x27;re basically asking if the car avoided doing something that would have been more destructive (e.g not hitting 2 people but only hitting one person instead). Or if there was no conflict decision making that happened and the car accidentally ran into the person and it resulted in a fatality.- AI decides between one life lost as a better potential outcome- AI made no decision and just happened to hurt someone

Please don&#x27;t, he&#x27;s going to launch into some philosophical dilemma that&#x27;s not relevant because these cars aren&#x27;t making decisions.

I think it&#x27;s a reference to the imaginary trolley problem, which is often raised as an objection to self-driving capability.The idea is that making a decision between killing one person vs killing something more valuable than one person (eg. multiple people) is something that drivers do on a regular basis. Computers may not be capable of correctly evaluating the ethics of that decision.

One of the most oft-discussed questions around self-driving cars is what should happen if your car can choose between two options: 1) An action that will have a better chance of protecting the people inside the car, 2) An action that will protect people outside of the car, with a whole bunch of variations.For example - What if someone steps out into the road and you can either hit them, or run into a wall? The former makes it far more likely the occupants of the car will be safe, the latter saves the life of a pedestrian at the expense of possible injury to people in the car.It is kind of an interesting question, because people can&#x27;t generally react quickly and rationally enough in a situation like that for it to matter, but in theory, a computer can. But it also dominates the discussion in a way that likely far overstates how frequently this will actually be a decision a computer has to make.I think that is what the person you&#x27;re replying to is asking. Given that this occurred at night and on (very likely) uncrowded streets, I feel like its far more likely the car (and backup driver) just somehow missed the pedestrian.

I believe they are asking whether that the vehicle struck pedestrian A in the course of avoiding something else like a pedestrian B.

I suspect they mean the trolley car problem: &quot;you are either going to crash into a group of school kids, or into a group of nuns. Who do you kill?&quot;I.e. did the uber car kill this 1 pedestrian, because it avoided killing 2 other people? Did the uber car opt for the &quot;least killing&quot; in this siutation?I think this whole trolley car question - while a nice philosophical question - is silly though. I dont think humans would do any better in an emergency situation. In that split second when you suddenly find yourself in the &quot;oh sh-&lt;IMPACT&gt;&quot; situation, do you have time to a) make a deliberate, reasoned &amp; fully-informed decision, and then b) control the vehicle effectively to follow through on that decision? I doubt it. If you had a second or two to think it through and then deliberately make a choice and steer towards that crowd of nuns, you&#x27;d probably be able to entirely avoid the accident anyway.More likely is you&#x27;d probably do what I am sure most people do which is gasp, stamp on the brakes, shut their eyes and hope for the best ... assuming you even had time to realise there was about to be a crash before it happened. Some people might swerve, but I suspect they do that instinctively to avoid something in their way, not as a decision to hit something else.I just dont think self-driving cars will get themselves into situations where they have to chose who to kill in the first place. And even if they did get into those situations, I really, really doubt it would be due their actions, and I&#x27;d certainly trust it to avoid the crash in the first place a whole lot more than the average human driver in the same scenario.Some could argue &quot;Ah but yes computers are so fast that they CAN make that informed decision about who to kill in milliseconds! So they have to make a choice! The question stands!&quot;. I&#x27;d argue back that in those milliseconds before the crash was inevitable, they&#x27;d act to prevent the accident before a human would even know what the hell was going on anyway. After that it just starts getting into a game of who can conjure up the most ludicrous hypothetical situation that rarely - if ever - happens in real life.

Forgive me, I don&#x27;t understand what you mean.  Please elaborate.

I&#x27;m pretty sure everyone involved and observing also vacillate between those two mindsets. Regardless, events like this will likely hamper progress.

Yeah, it will be interesting to see if this was an intentional (algorithmic) response or a failure of sensors somewhere.I go back and forth with thinking the algorithms in things like self-driving cars should be public. Anything which must make a choice such as smash into bus or smash into bicycle — the public should be aware of the choices the company chooses to implement. Obviously this example is simplistic but it illustrates my point.Today I’m definitely leaning towards it should be public and scrutinized. Next month I may be worried this would hamper progress.

On purpose to protect something &quot;more valuable,&quot; or on accident?

In the Tesla one the sensors did not recognize the white trailer of an 18 wheeler against the sky from news reports.

IIRC in that case the driver T-boned a truck crossing the road, because the side of the truck was so shiny that it read as the horizon to the car&#x27;s sensors.

Hm, there have been other crashes where the &#x27;driver&#x27; in the car died (I recall one where a car in autonomous mode rear-ended a truck, something about how the sensor was blinded by the setting sun? And it had alerted the driver to take control several times before the crash) but this is the first incident I can recall where a pedestrian was struck and killed by an autonomous vehicle.

I believe it&#x27;s the first &quot;third-party&quot; death.

Is this the first death ever?(Do we count Tesla&#x27;s autopilot as self-driving?)

Which is why the machine should be backup for the always-driving human, only leaping in to correct failings.

&gt; Human beings simply cannot switch between &quot;not focussed&quot; and &quot;in charge of a carThis is not how many people drive in non-autonomous cars? Commuting is on auto-pilot for many of us; you don&#x27;t really remember how you got to work and I see people tying their shoelaces, calling, eating sandwiches, reading the paper, chatting on whatsapp all the time. Sure it&#x27;s all illegal (in a lot of countries), but people are bored and drove that road 1000x.

I agree with what you are saying, and that it would be very dangerous to sell cars where drivers may be lulled into complacency of thinking they don&#x27;t need to pay attention to the road, until self driving cars get good enough that human attention really isn&#x27;t needed.However, these are test vehicles. The driver&#x27;s full-time job is to be focused on the road and what the car is doing. They shouldn&#x27;t have any misconceptions about the need for them to stay focused on the road. Now granted, even then attention will lapse occasionally, just like it does when you are actually driving. But I don&#x27;t think that being physically in control of the wheel is necessary for maintaining focus at the same level as a good driver. Driver&#x27;s Ed instructors need this skill. Also some back seat drivers I know are quite good at maintaining strong focus on the road regardless of who is driving :)

How about autonomous on the highway, turnpikes, etc. And human-operated on streets where a car might ever need to yield to pedestrians or cyclists outside of red-lights.

Certainly, humans will never be perfect but there is the wicked problem of how we interact with systems that behave correctly 99% of the time but require human attention the other 1%. There was an excellent Econ Talk podcast about this topic a while ago:<a href="http:&#x2F;&#x2F;www.econtalk.org&#x2F;archives&#x2F;2015&#x2F;11&#x2F;david_mindell_o.html" rel="nofollow">http:&#x2F;&#x2F;www.econtalk.org&#x2F;archives&#x2F;2015&#x2F;11&#x2F;david_mindell_o.htm...</a>However, in my opinion, the driver and by extension his employer Uber must be held responsible in this case. Uber have a license to test their self-driving cars only if there is a human driver in control at all times. Regardless of the crazy things their software decides to do, the human is there as a person of ultimate legal liability.

Human beings simply cannot switch between &quot;not focussed&quot; and &quot;in charge of a car, taken out of autonomous mode, and actively avoiding collision&quot; fast enough to avoid most accidents, unfortunately. Neither can humans maintain the focus required to be ready to do that when 99.9% of the time they&#x27;re not required to do anything.Semi-autonomous cars have drawbacks.

Given these cars have been blowing red lights, the human drivers aren&#x27;t doing their jobs and overriding the car.

The only way this is a reasonable failure of both the computer and the human driver is if they both physically had no time to react. Maybe that was the case, maybe it wasn&#x27;t.

Motor vehicle occupants are much more likely to survive than pedestrians and cyclists are. But blaming victims (especially dead ones) has a long and storied history as far as law enforcement and the road.Hopefully with self driving vehicles we can start to move beyond this attitude, especially when there should be ample camera footage available to help dispel the usual claims that the person struck &quot;came out of nowhere&quot; or whatever.

&gt; I&#x27;m very curious to see how they&#x27;ll investigate this and who will be determined to be at fault (person behind the wheel or Uber). It will likely set a precedent.It&#x27;s also plausible that the pedestrian was at fault. Under some circumstances it would be impossible for a human driver to avoid collision with a pedestrian, this may be no different.

Concioussness requires a body. This is how it is, here.

Nothing I love more than HackerNews armchair engineering.It&#x27;s not that simple, you&#x27;re assuming the car even had some indication that something was wrong. For all we know the car&#x27;s vision was showing high confidence it saw an open road.

Car rapidly approaches you from behind.Slow down?

&quot; Obviously more information is needed, but I thought the entire point of having a driver behind the wheel is to manually intervene to prevent this very situation?&quot;I think it&#x27;s time to point out the obvious, and require that autonomous cars apply the brakes first, and THEN require driver intervention.And that they be a whole lot quicker to err on the side of braking.Cameras getting fuzzy ?Slow down.Your ML algorithms are showing lower confidence measures for how they classify nearby objects and trajectories ?Slow down.nearby vehicles slowing down and you don&#x27;t know why?Slow down.This is inexcusable.

It could have been the fault of the pedestrian. I&#x27;m not saying it is, just saying there is a third possibility.

If humans can&#x27;t prevent themselves from having accidents 100% of the time when they&#x27;re in total control of the vehicle, why should we expect they could prevent an accident 100% of the time when they&#x27;re in partial control of the vehicle?

Obviously more information is needed, but I thought the entire point of having a driver behind the wheel is to manually intervene to prevent this very situation?I&#x27;m very curious to see how they&#x27;ll investigate this and who will be determined to be at fault (person behind the wheel or Uber). It will likely set a precedent.

Yeah. Their motto seems to be &#x27;easier to ask forgiveness&#x2F;pay fines instead of asking for permission&#x27;.

&gt; Uber&#x27;s self-driving program seems to be one of the most unethical, most poorly-run in the entire industry.I mean, it&#x27;s Uber. I&#x27;m not sure exactly what people were expecting. Their entire schtick is being unethical and poorly-run.

We do not yet know whether another autonomous vehicle or human for would have done better in this situation. THAT SAID, asshole company fields asshole autonomous driver. Did we expect differently? Uber cuts every corner.

Yup - I&#x27;ve personally witnessed one of these almost zoom through a red light at an intersection and was appalled. I&#x27;m honestly not surprised...in their effort to race to the bottom Uber&#x27;s autonomous vehicle program has been extremely aggressive - to the point of total disregard for safety. Their business model may be built on pipe dreams of autonomous travel (without drivers to pay), and hopefully this shutters that. In fact, the whole company should probably be shut down IMO - but I digress.

EDIT: Yeah this was a bit reactionary on my part. I get it, it&#x27;s not over the $150.It tells a lot about Uber that they weren&#x27;t willing to fork over $150 for a permit.

IIRC, Uber initially launched its self driving cars by flouting California law: <a href="https:&#x2F;&#x2F;www.google.com&#x2F;amp&#x2F;s&#x2F;www.cnet.com&#x2F;google-amp&#x2F;news&#x2F;ubers-self-driving-cars-back-on-sf-streets-but-now-theyre-legal&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.google.com&#x2F;amp&#x2F;s&#x2F;www.cnet.com&#x2F;google-amp&#x2F;news&#x2F;ub...</a>

Agreed. I&#x27;ve been in the cars in Pittsburg and the operators laughed about how poorly it performed on certain roads. Slanted telephone polls caused the vehicle to literally slalom down the road.

Calls to wait for facts never stopped the internet outrage machine in the early moments of news release. I doubt it will do much here, sadly. But it&#x27;s worth pushing back against, even if it won&#x27;t work IMO.People will always relish in their biases being confirmed... even after a rare one-off event with a sample size of 1 in a burgeoning industry.

We should wait for the results of investigation first. It could easily happen the killed person did some bonehead move that resulted in unavoidable crash. I was once told how a friend of mine fatally hit an older man that decided to cross the highway quickly, giving him no chance to react and a life-long trauma.

Not sure how that regulation would have made their cars safer in a way that&#x27;s relevant to this accident?Local governments have stopped other self-driving companies before with cease-and-desist letters with threats of serious punitive fines. I highly doubt by not buying a $150 license Uber would suddenly have carte-blanche to do what they want, that&#x27;s not how it works.If the state gov was really interested in stopping them they easily could have regardless.

Their San Francisco testing program didn&#x27;t even have a license to test because they didn&#x27;t want to pay for a permit.<a href="https:&#x2F;&#x2F;www.theverge.com&#x2F;2017&#x2F;2&#x2F;27&#x2F;14698902&#x2F;uber-self-driving-san-francisco-dmv-email-levandowski" rel="nofollow">https:&#x2F;&#x2F;www.theverge.com&#x2F;2017&#x2F;2&#x2F;27&#x2F;14698902&#x2F;uber-self-drivin...</a>Nothing to revoke if they never had a license to begin with!

Fully agreed regarding Uber&#x27;s ethics, but...&gt; Waymo et al has had none of these issues.... absence of negative media coverage doesn&#x27;t prove anything.

You&#x27;d think Uber would be extra-vetted before they were allowed to do something like this.

&gt; Waymo et al has had none of these issues. Time to revoke their licence to test?I think Alphabet is to blame here too in a sense (or at least the industry as a whole). Why didn&#x27;t they push for more strict testing rules? It&#x27;s in their interest too.

This is how cars came to be on the road in the first place.  Despite the fact that they were erratic and deadly, and spewed noxious gas everywhere, the manufacturers and automobiling clubs succeeded in convincing everybody that roads were for cars.  Uber may well be the worst actor in the self-driving industry, but the industry as a whole is going to need to invent new crimes akin to jaywalking.   This will allow it to shift blame to those killed by autonomous vehicles, just as jaywalking implies pedestrians are to blame for being hit by manned vehicles.

Yeah I missed that but it doesn&#x27;t look good [1] (december 2016).[1] - <a href="https:&#x2F;&#x2F;www.theguardian.com&#x2F;technology&#x2F;2016&#x2F;dec&#x2F;19&#x2F;uber-self-driving-cars-bike-lanes-safety-san-francisco" rel="nofollow">https:&#x2F;&#x2F;www.theguardian.com&#x2F;technology&#x2F;2016&#x2F;dec&#x2F;19&#x2F;uber-self...</a>

Uber&#x27;s self-driving program seems to be one of the most unethical, most poorly-run in the entire industry. There were plenty of reports before warning us of the questionable quality of their research: cars changing lanes suddenly, blowing red lights, and now one finally killed someone. Every prior incident should have been treated as a serious matter.Waymo et al has had none of these issues. Time to revoke their licence to test?

Fine&#x2F;sue the entity that made the algorithm, just like every other situation like that. Or, if you can prove negligence on the part of someone inside that company, prosecute them.

&gt;What do you do with an alogrithm to provide justice?Assuming Arizona hasn&#x27;t produced law specifically for fatalities in their public road self driving test program... The algorithms are just one detail of the total system design, and the design of the system isn&#x27;t really the issue. People don&#x27;t get killed by designs or code.The verification of it&#x27;s safety, and the decision to deploy that system on uncontrolled public streets will be the issue. People made these decisions, not an algorithm.

yes, but with a human driver you have someone to take responsibility &#x2F; punish for, what do you do with an alogrithm to provide justice?

I always take the &quot;outside the crosswalk&quot; reports with a grain of salt.  Hereabouts, if the pedestrian lands outside the crosswalk, they&#x27;re assumed to have been crossing outside of it, even if later eyewitness reports and video footage show they were in the crosswalk and the car knocked them out of it.

It appears the woman was not using the crosswalk, but really that doesn&#x27;t make much difference in the reaction to the report.  Uber&#x27;s self driving tech seems a bit under-baked at the moment.

If that&#x27;s something the software can&#x27;t handle well then it&#x27;s going to be a huge problem in many parts of the world. In many places in Europe, Africa and Asia for instance pedestrians will cross anywhere and everywhere at any time.

Does it really need to be said that an autonomous vehicle still needs to manage to not kill people regardless of their position relative to any nearby crosswalks?Today you can cross in the middle of the street and take a calculated risk that the people approaching down the street will stop or slow down enough to not kill you. If they go ahead and kill you anyway, they&#x27;ll still be prosecuted because their attention should be on the road, and the most basic assumption behind their driver&#x27;s license is that this person is competent enough and fit enough to drive what amounts to a deadly weapon without killing the people around them. When you violate that assumption, the fault is probably on you. Obviously there are exceptions, such as when somebody intentionally jumps in front of your vehicle, but crossing outside of the crosswalk to get across the street is not even close to the same level as trying to commit suicide and fault will be found accordingly.

Doesn&#x27;t mitigate the tragedy or implications for autonomy, but per the article, the pedestrian was outside the crosswalk.

I don’t know the number of hours their system has been driving but I highly doubt it is enough to draw comparisons to the average Arizona driver who hits 0 pedestrians. I also don’t think this works like this in the court of public opinion anyways. 
This system can never make a mistake of this magnitude and be publically accepted.

&gt;The reality though is how many pedestrians are hit by manual driver&#x27;s vs automous drivers? On a percentage basis I gotta believe autonomous cars are orders of magnitude safer.That might be a fair question, except that one death at this early stage of limited realistic trials makes me doubt the correctness of your belief. On the statistics that we have right now, I&#x27;m not seeing how autonomous cars are safer.&gt;Self driving cars aren&#x27;t going to be perfect.Shouldn&#x27;t self-driving cars be held to a better standard rather than compared generously to objectively terrible existing standards? That we think of pedestrians regularly killed in &quot;accidents&quot; as acceptable collateral is already pretty horrific.

Our society has evolved quite a bit from the times of the Industrial Revolution. It’s possible for progress to be made while having life-saving regulations.

what about tesla though? by november 2016 they&#x27;ve accumulated 1.3 billion self droven miles [1]. autopilot was released in september 2014, so that&#x27;s 2 years, or 600 million miles per year. and that was ramping up during those years, so i think it&#x27;s safe to assume at least 1 billion miles &#x2F; year these days.2 deaths in over 2 billion miles is roughly 10 times safer than human drivers.[1] <a href="https:&#x2F;&#x2F;electrek.co&#x2F;2016&#x2F;11&#x2F;13&#x2F;tesla-autopilot-billion-miles-data-self-driving-program&#x2F;" rel="nofollow">https:&#x2F;&#x2F;electrek.co&#x2F;2016&#x2F;11&#x2F;13&#x2F;tesla-autopilot-billion-miles...</a>

This is the important statistic. Evidence so far points to autonomous drivers being worse than human drivers. In which they definitely have no place on the streets without a lot more R&amp;D.

&gt; On a percentage basis I gotta believe autonomous cars are orders of magnitude safer.Why?NHTSA reports 1.15 fatalities per 100 million vehicle miles travelled in 2015.Waymo advertises &gt;5 million road miles travelled. Let&#x27;s say Waymo + Uber have driven 10 million vehicle miles and killed 1 person. That makes them 10 times more dangerous than human driven cars.<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Motor_vehicle_fatality_rate_in_U.S._by_year" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Motor_vehicle_fatality_rate_in...</a>

The problem is that you &#x27;believe&#x27; it to be orders of magnitude safer.How about we test it properly before allowing it on public roads?

Striking a pedestrian in a crosswalk is horrible and terrible news. The reality though is how many pedestrians are hit by manual driver&#x27;s vs automous drivers? On a percentage basis I gotta believe autonomous cars are orders of magnitude safer. Self driving cars aren&#x27;t going to be perfect. How many people lost their lives in early factories to machines during the industral revolution? Imagine if they had pulled the plugs back then.

It&#x27;s not &quot;intelligent&quot;, it&#x27;s a dumb computer that is programmed by humans. The people responsible for putting it on the road must be liable.

&gt; may as well declare the AI guilty now right?AI (short of AGI) is never guilty; it&#x27;s creators and operators, OTOH...

They will have video for sure if since they are testing, then we will see. Your statement about reaction time is assuming so many things, may as well declare the AI guilty now right?

This is Uber, who designed systems for remotely wiping entire offices when they thought that the data might be called for by the authorities.Perhaps there will be unexpected &#x27;issues&#x27; meaning that such data will have been lost in this case.

I imagine these vehicles (especially test vehicles) are also recording regular video and therefore getting a clear picture of what happened should be straight forward.

That isn&#x27;t correct. Culture is learning from other people&#x27;s mistakes. The only question is whether the human can and is willing to accept the lessons (and whether the lessons are correct).

Uber&#x27;s algorithms might learn from this. None of the other self driving car companies will.

If there&#x27;s one company who hasn&#x27;t demonstrated it&#x27;s learned from its mistakes it&#x27;s Uber.But let&#x27;s extrapolate. Say one day there are 20 self driving car companies. Should they be required to share what they learn so the same mistakes aren&#x27;t repeated by each company or does the competitive advantage outweigh the public benefit from this type of information sharing?

Airlines are the same way. Every time there’s been a crash they’ve learned something and changed procedures to prevent it again. It’s made flying pretty safe, unless you’re an animal flying on United...

I agree completely. One thing that is important to me is that the whole self-driving field will learn from every mistake. In other words, every self-driving car should only make the same mistake once, whereas with humans, each human has to learn only from their own mistakes.

&gt; I have seen no evidence that any autonomous vehicle currently deployed can react faster than an alert and aware human.The argument I&#x27;ve always heard is that an autonomous systems will outperform humans mostly by being more vigilant (not getting distracted, sleepy, etc.) rather than using detectors with superhuman reaction times. Obviously, whether or not this outweighs the frequency of situations where the autonomous system gets confused when a human would not is an empirical question that will change as the tech improves.

&quot;Alert and aware human&quot; is already a high standard, given how most humans drive in routine mode, which is well understood to be much worse than &quot;alert and aware&quot;.From what I&#x27;ve seen I wouldn&#x27;t trust autonomous cars to &quot;understand&quot; all situations. I would trust Waymo cars to understand enough to avoid hitting anything (at a level better than a human), at the risk of being rear-ended more often. Everything I&#x27;ve seen from Tesla and Uber has given me significantly less confidence than that.

I live and commute in Waymo country, and see evidence of quick reactions, though I can&#x27;t say for sure whether it&#x27;s an alert human taking over. Mostly, the Waymo vehicles still drive conservatively.

Not a truly autonomous vehicle example but this is a case where most likely the car reacted before the driver was even aware of a problem: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=APnN2mClkmk" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=APnN2mClkmk</a>I agree with the sentiment though. This has been a major selling point for this technology, but it has not been sufficiently demonstrated yet.

Consider that autonomous braking systems know the distance required to stop and activate appropriately. Try getting a human to mimic that.

I have seen no evidence that any autonomous vehicle currently deployed can react faster than an alert and aware human.  The commentariat tends to imagine that they can, and it&#x27;s certainly plausible that they may eventually be.  But I&#x27;ve never seen anyone official claim it, and the cars certainly don&#x27;t drive as though they can quickly understand and respond to situations.

&gt; don&#x27;t conclude that humans aren&#x27;t ready to be driving?I&#x27;m not sure I see anyone here making that conclusion. I think you&#x27;re the only one who&#x27;s brought it up.For one, I personally sure as fuck do not think humans are ready to be driving.&gt; That&#x27;s an invalid conclusion to draw from this accident.The conclusion was not invalid at all. Other self-driving car companies have driven more miles than Uber has and they have done so safely. Uber has even taken its cars off the road, so even Uber agrees that their self-driving cars are not ready for the roads yet.It is also important to take into account what Uber is like when it comes to safety and responsibility. They have knowingly hired convicted rapists for drivers, they have ridiculed and mocked and attempted to discredit their paying customers who have been raped by their employees&#x2F;drivers, they have spied on journalists, they have completely disregarded human safety on numerous accounts. A company with a track record like Uber&#x27;s probably not should be granted a license to experiment with technology like these self-driving cars on public roads.They just aren&#x27;t responsible enough.

1 July 2016 &quot;Tesla driver dies in first fatal autonomous car crash in US&quot;<a href="https:&#x2F;&#x2F;www.newscientist.com&#x2F;article&#x2F;2095740-tesla-driver-dies-in-first-fatal-autonomous-car-crash-in-us&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.newscientist.com&#x2F;article&#x2F;2095740-tesla-driver-di...</a>

He&#x27;s arguing from foundations. The hypothesis is that autonomous drivers should perform better at task X than the null hypothesis (aka human drivers). So any instances where autonomous drivers do not seem to perform better are all potential counter-arguments to that hypothesis.The fact that human drivers aren&#x27;t particularly good isn&#x27;t really relevant, beyond setting a correspondingly low bar within the null hypothesis.This all ties to regulation allowing these vehicles to drive on public roads, because that regulation was permitted due to the above hypothesis and hopeful expectations that it would be true.Obviously, I haven&#x27;t seen the entirety of the data set to know fatalities per car-mile. Which would be the relevant statistic here. I also didn&#x27;t see such a number in your post, which I&#x27;m assuming means you are probably not aware either. But simply providing the numbers for the null hypothesis doesn&#x27;t do anything.

you have to take into account miles driven. Yes, we have 37K fatalities, but trillions of miles driven.  So, it comes out something like 1.2 fatalities per 100 million miles driven for human drivers.  Which means, so far, self driving cars have a worse record per 100 million miles driven.  So far.

If solution does not solve anything, what&#x27;s the point of the „solution“?
Most of the deaths on the road I hear in my country are the result of someone doing something really reckless and stupid. „Normal“ drivers do not kill themselves or others. So if self-driving car is only as good as a dumb driver — I do not want these cars on the road.
Also, interetingly enough there is little talk about human driver assisted by technology, not replaced by it. For some reason it is binary: either humand driver or self-driving. How about some human drivers plus collsion avoidance system, infrared sensing system (way too many people die there simply becaue they walk on the road in the dark without any reflectors&#x2F;lights), etc.

As of late 2017, Waymo&#x2F;Google reported having driven a total of 4 million miles on roads. Given that Waymo is one of the biggest players, it&#x27;s hard to see how all of autonomous cars have driven 100 million miles at this point.<a href="https:&#x2F;&#x2F;techcrunch.com&#x2F;2017&#x2F;11&#x2F;27&#x2F;waymo-racks-up-4-million-self-driven-miles&#x2F;" rel="nofollow">https:&#x2F;&#x2F;techcrunch.com&#x2F;2017&#x2F;11&#x2F;27&#x2F;waymo-racks-up-4-million-s...</a>Nevermind that the initial rollouts almost always involve locales in which driving conditions are relatively safe (e.g. nice weather and infrastructure). Nevermind that the vast majority of autonomous testing so far has involved the presence of a human driver, and that there have been hundreds of disengagements:<a href="https:&#x2F;&#x2F;www.theverge.com&#x2F;2018&#x2F;1&#x2F;31&#x2F;16956902&#x2F;california-dmv-self-driving-car-disengagement-2017" rel="nofollow">https:&#x2F;&#x2F;www.theverge.com&#x2F;2018&#x2F;1&#x2F;31&#x2F;16956902&#x2F;california-dmv-s...</a>Humans may be terrible at driving, but the stats are far from being in favor of autonomous vehicles. Which is to be expected given the early stage of the tech.

Actually there was a metric of accident per distance drived. It was often pull out by Tesla before their first fatal accident last year (IA didn’t saw rear of truck because of sun or smthg like that).This metric was often decried because it had poor statistical significance. It would however be nice to update this metric according to this death.Maybe now this metric would indicate that IA are more dangerous than human, will be interesting if the perception of this flawed metric will evolve in reaction...

&gt; That&#x27;s a very bad look; the whole point of self-driving cars is that they can react to unexpected circumstances much more quickly than a human operator, such as someone walking out into the road. Sounds like Uber&#x27;s platform may not be up to that standard yet, which makes me wonder why they&#x27;re on public roads.That&#x27;s an invalid conclusion to draw from this accident.There were 34,439 fatal motor vehicle crashes in the United States in 2016 in which 37,461 deaths occurred. This resulted in 11.6 deaths per 100,000 people and 1.16 deaths per 100 million miles traveled.[1] As far as I know, the instance this article is about is the first death in an autonomous vehicle accident, meaning that all the 2016 accidents were humans driving.Why is it that you see one death from an autonomous car and conclude that autonomous cars aren&#x27;t ready to be driving, but you see 37,461 deaths from human drivers and don&#x27;t conclude that humans aren&#x27;t ready to be driving?I admit that there just aren&#x27;t enough autonomous cars on the road to prove conclusively that autonomous cars are safer than human-operated cars at this point. But there&#x27;s absolutely no statistical evidence I know of that indicates the opposite.[1] <a href="http:&#x2F;&#x2F;www.iihs.org&#x2F;iihs&#x2F;topics&#x2F;t&#x2F;general-statistics&#x2F;fatalityfacts&#x2F;state-by-state-overview" rel="nofollow">http:&#x2F;&#x2F;www.iihs.org&#x2F;iihs&#x2F;topics&#x2F;t&#x2F;general-statistics&#x2F;fatalit...</a>EDIT: Let&#x27;s be clear here: I&#x27;m not saying autonomous cars are safer than human drivers. I&#x27;m saying that one can&#x27;t simply look at one death caused by an autonomous car and conclude that autonomous cars are less safe.

That&#x27;s a really good point that many (myself included) aren&#x27;t mentioning. Stopping distance is a universal, reaction times be damned. Would be curious to see if that played a part.

Not bad even a self driving car can stop in time for deer jumping out from behind tree&#x2F;bush cover.People still need to look both ways when coming out from behind cars.I want more details.

Yet this tragic accident will now set a precedent and finally start to scratch the delicate unspoken question:
Who is responsible now?

The cars telemetry data will undoubtedly point to a unavoidable tragedy if it&#x27;s just uber analyzing it.

If you have good judgement, it&#x27;s quite easy to prevent a fatality. I.e. I go slower in places where I know pedestrians might jump out.

I’m kind of in the Elon Musk camp here where you gotta break some eggs to make an omelette? Human-driven cars kill a lot of pedestrians today, but we can actually do something to improve the human-recognition algorithms in a self-driving car.As long as self driving cars represent an improvement over human drivers, I’m ok with them having a non-zero accident date while we work out the kinks.

you seem to be giving uber a big benefit of the doubt. these autonomous cars generally go slow. tempe has flat roads with great line of sight and clear weather. coefficient of friction? highly doubt it. the sensors should be looking more than just straight a head

It&#x27;s not unreasonable for there to an expectation of basically zero accidents of this nature during testing in cities. The public puts a huge amount of trust in private companies when they do this. And, pragmatically, Google, Uber, etc, all know that it would be horrible publicity for something like this to happen. One would think they&#x27;d be overly cautious and conservative to avoid even the possibility of this.Lastly, the whole point of the human operator is to be the final safety check.You&#x27;re right that we have no idea of the cause until the data is analyzed (and the human operator interviewed). Yet, my first thought was, &quot;Of course it&#x27;d be Uber.&quot;

&gt;If a pedestrian suddenly rushes out into the street or a cyclist swerves into your path, the deciding factor is often simply the coefficient of friction between your tyres and the road.This only true for the uninitiated, never let it get to that. I drove for Uber&#x2F;Lyft&#x2F;Via in New York city so I can experience and study these situation. These sort of accidents are preventable. The following is the basic:1.) Drive much slower in areas where a pedestrian and cyclist can suddenly cross your past.2.) Know the danger zone. Before people &quot;jump into traffic&quot; or a cyclist swerve in front of you, they have to get into position, this position is the danger zone.3.) Extra diligence in surveying the area&#x2F;danger zone to predict a potential accident.4.) Make up for the reduce speed by using highways and parkways as much as possible.It helps that Manhattan street traffic tends to be very slow to begin with. Ideally I will like use my knowledge to offer a service to help train autonomous vehicles to deal with these situation. It has to be simulated numerous times in a closed circuit for the machine to learn what I&#x27;ve learn intuitively driving professionally in NYC.

Agreed that it&#x27;s too early to really say one way or another. In maybe 100k miles of urban driving I&#x27;ve had one cyclist run into my car and a girl on her phone walk directly into the corner of the front, I was at a complete stop watching them both times.Until there&#x27;s a detailed report it&#x27;s really hard to say if it was preventable or not - but I think regardless the optics are bad and this is going to chill a lot of people&#x27;s feelings on self driving whether or not that is an emotion backed by data.

&gt; If a pedestrian suddenly rushes out into the street or a cyclist swerves into your path, the deciding factor is often simply the coefficient of friction between your tyres and the road.This holds iff the control input you apply is only braking. Changing the steering angle is generally far more effective for the &quot;pedestrian darts out from hidden place onto road&quot; situation. It&#x27;s far better to sharply swerve away to ensure that there&#x27;s no way the pedestrian can get into your path before your car arrive there than it is to stand on the brakes and hope for the best.Indeed, the faster you&#x27;re moving, the more you should consider swerving away over braking -- take advantage of that lethal speed to clear the pedestrian&#x27;s potential paths before he can get into yours.Yes, this intentionally violates the letter of the traffic laws (and might involve colliding into a a parked or moving automobile on the other side of the road) and also involves potentially unusual manoeuvring on a very short deadline; but it&#x27;s far better to avoid a vehicle-pedestrian collision even if it&#x27;s at the cost of possibly busting into the opposing lane &#x2F; driving off the road &#x2F; hitting a parked car. Decently experienced drivers can do this, I can do this (and have successfully avoided a collision with a pedestrian who ran out between parked cars on a dark and raining night), there&#x27;s no fundamental reason that computer-controlled cars can&#x27;t do this.

That’s technically true but the number of truly unavoidable cases is orders of magnitude lower. With human drivers, “jumped out in front of me” really means inattention 99.9% of the time and police departments historically tend to be unlikely to question such claims. (For example, here in DC there have been a number of cases where that was used to declare a cyclist at fault only to have private video footage show the opposite - which mattered a lot for medical insurance claims)With self-driving cars this really seems to call for mandatory investigations by a third-party with access to the raw telemetry data. There’s just too much incentive for a company to say they weren’t at fault otherwise.

The hope is that AV can see 360 and observe things outside of blind spots further away.  So the kid running from a yard into the road after a ball should be safer, but a person walking out from behind a parked truck wouldn&#x27;t.

If a pedestrian suddenly rushes out into the street or a cyclist swerves into your path, the deciding factor is often simply the coefficient of friction between your tyres and the road.I mentioned in another comment that something I use to try to improve my own driving is watching videos from &#x2F;r&#x2F;roadcam on reddit, and trying to guess where the unexpected vehicle or pedestrian is going to come from.Here&#x27;s an example of a pedestrian suddenly appearing from between stopped cars (and coming from a traffic lane, not from a sidewalk), and a human driver spotting it and safely stopping:<a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=wYvKPMaz9rI" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=wYvKPMaz9rI</a>Why can&#x27;t a self-driving car do this?

&gt; That&#x27;s a very bad look; the whole point of self-driving cars is that they can react to unexpected circumstances much more quickly than a human operator, such as when someone walks out into the road.Some of these accidents are unpreventable by the (autonomous) driver. If a pedestrian suddenly rushes out into the street or a cyclist swerves into your path, the deciding factor is often simply the coefficient of friction between your tyres and the road.The autonomous vehicle and the human attendant might have made a glaring error, or they might have done everything correctly and still failed to prevent a fatality. It&#x27;s far too early to say. It&#x27;s undoubtedly a dent to the public image of autonomous vehicles, but hopefully the car&#x27;s telemetry data will reveal whether this was a case of error, negligence or unavoidable tragedy.

The first martyr of the AI age, I suppose.

The bicyclist did not &quot;suddenly&quot; jump out in front. The Uber car did not sense the bicyclist.

I&#x27;ve been saying this from the beginning: it&#x27;s not enough for self-driving cars to be &quot;better than the average driver&quot;. They need to be at least 10x better than the best drivers.I find it crazy that so many people think it is. First off, by definition like 49% of the drivers are technically better than the &quot;average driver&quot;.Second, just like with &quot;human-like translation&quot; from machines, errors made by machines tend to be very different than errors made by humans. Perhaps self-driving cars can never cause a &quot;drunken accident&quot;, but they could cause accidents that would almost never happen with most drivers.Third, and perhaps the most important to hear by fans of this &quot;better than average driver&quot; idea, is that self-driving cars are not going to take off if they &quot;only&quot; kill almost as many people as humans do. If you want self-driving cars to be successful then you should be demanding from the carmakers to make these systems flawless.

Could a human have reacted fast enough to stop for someone jumping out in front of them? If the person jumped out so fast that nobody could have possibly reacted in time, then it&#x27;s not a stain on the technology- even with an instantaneous application of brakes, a car still takes a while to come to a stop. If the human jumped out ten seconds earlier and was waving her hands for help, then it&#x27;s an issue.&quot;the whole point of self-driving cars is that they can react to unexpected circumstances much more quickly than a human&quot;This statement isn&#x27;t really true. Do you think Uber is investing in this because it makes their passengers safer? No, they are pretty much immune to any car crashes with a human driver, the risk and insurance hit is assumed by the driver. They are doing this to save money. They don&#x27;t have to pay a driver for each trip. The safety aspect is just a perk, but Uber will be pushing this full force as long as the self-driving cars are good enough.I&#x27;m more concerned that Uber will &quot;lose&quot; the video evidence showing what kind of situation it was, and we&#x27;ll never be able to know if a human would have had ample time to react.

With sufficient data, I&#x27;d expect self-driving cars to be better at predicting what such leans mean. Moreover, for every one human driver who notices such a lean, there may be another human driver that doesn&#x27;t even notice a pedestrian who has already started walking.

Are you sure that supervised learning does not create the same classification capabilities in self driving car AIs?

They also fail to &quot;write&quot; human body language. Nobody else can predict what the autonomous vehicle will do.It gets worse when a person is sitting in what appears to be the driver&#x27;s seat. If the car is stopped and that person is looking toward another passenger or down at a phone, nobody will expect the vehicle to begin moving. Making eye contact with the person in that seat is meaningless, but people will infer meaning.

Humans also can&#x27;t read humans&#x27; body language. A pedestrian waiting at a corner isn&#x27;t waiting for the weather to change. They are waiting for passing cars to stop, as required by law. But passing cars speed by instead of stopping, unless the pedestrian does a lunge into the street -- preferable a bluff lunge, since most drivers still won&#x27;t stop, preferring to race the pedestrian to the middle of the lane.

OT, but I would love to see how self-driving AIs handle something like Vietnam moped traffic and pedestrian crossings. The standard behavior for pedestrians is to walk slowly and keep walking -- stopping, even if it seems necessary to avoid a speeding driver, can be very dangerous, as generally, all of the street traffic is expecting you to go on your continuous path. It&#x27;s less about reading body language than expectation of the status quo:<a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=nKPbl3tRf_U" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=nKPbl3tRf_U</a>

I walk to work every day, and I can assure you that most human drivers have zero awareness of what pedestrians are about to do.Even on crosswalks. If I just strolled out on a crosswalk without looking, and waiting for drivers who were paying no attention, I’d be long dead.

The main problem with self-driving cars is that they can&#x27;t &quot;read&quot; humans&#x27; body language.  A human driver can see pedestrians and cyclists (and other cars) and have a rough idea of what they&#x27;re likely to do in the next few seconds, i.e. the pedestrian leaning out on the crosswalk curb is likely to step into the road soon.  Even a reaction time of milliseconds can&#x27;t make up for the (currently exclusive) human ability to read other humans and prepare accordingly.

Here&#x27;s a times article you reminded me of that mentions Saab engineers putting on crash helmets and running the 9-5 into a concrete moose.<a href="http:&#x2F;&#x2F;www.nytimes.com&#x2F;1998&#x2F;04&#x2F;05&#x2F;automobiles&#x2F;behind-the-wheel-1999-saab-9-5-flagship-s-new-thrust-hot-turbo-cool-seats.html" rel="nofollow">http:&#x2F;&#x2F;www.nytimes.com&#x2F;1998&#x2F;04&#x2F;05&#x2F;automobiles&#x2F;behind-the-whe...</a>

&gt; Sounds like Uber&#x27;s platform may not be up to that standard yet, which makes me wonder why they&#x27;re on public roads.Who exactly do you think is granting Google, Uber, etc. approval for trials like this on public roads? It&#x27;s going to be some bureocrat with zero ability to gauge what sort of safety standard these companies&#x27; projects have reached.There are no standards here... what were you expecting would happen?

It&#x27;s an interesting dynamic.  We want this tech to be much better than us terrible humans before we deploy it and anything like us terrible humans is not acceptable.

I&#x27;d also add the goal of self-driving cars is to decrease costs for ride-sharing, home-delivery companies, etc, and also to decrease congestion via coordination amongst autonomous vehicles.

&gt; Sounds like Uber&#x27;s platform may not be up to that standard yet, which makes me wonder why they&#x27;re on public roads.Uber cutting corners and playing fast and loose with legislation? Unheard of!Here&#x27;s hoping they get hit with a massive, massive wrongful death lawsuit.

That report ends by saying essentially, &quot;it may not be possible to prove the safety of self-driving cars&quot;. [1] So the value here is questionable and the same logic could apply to anything with a low frequency of occurrence. The value of air bags by this measure was not proven until they were already mandated.[1] &quot;even with these methods, it may not be possible
to establish the safety of autonomous vehicles prior to making them available for public use&quot;

&gt;&gt; Sounds like Uber&#x27;s platform may not be up to that standard yet, which makes me wonder why they&#x27;re on public roads.They&#x27;re on public roads because they&#x27;ve decided that deploying their tech is more important than safety.I don&#x27;t see any other explanation. We know that it&#x27;s pretty much impossible to prove the safety of autonomous vehicles by driving them, so Uber (and almost everyone else) have decided that, well, they don&#x27;t care. They&#x27;ll deploy them anyway.How do we know that? The report by RAND corporation:<a href="https:&#x2F;&#x2F;www.rand.org&#x2F;pubs&#x2F;research_reports&#x2F;RR1478.html" rel="nofollow">https:&#x2F;&#x2F;www.rand.org&#x2F;pubs&#x2F;research_reports&#x2F;RR1478.html</a><pre><code> Key Findings

 * Autonomous vehicles would have to be driven hundreds 
 of millions of miles and sometimes hundreds of billions 
 of miles to demonstrate their reliability in terms of 
 fatalities and injuries.

 * Under even aggressive testing assumptions, existing 
 fleets would take tens and sometimes hundreds of years 
 to drive these miles — an impossible proposition if the 
 aim is to demonstrate their performance prior to 
 releasing them on the roads for consumer use.

 * Therefore, at least for fatalities and injuries, 
 test- driving alone cannot provide sufficient evidence 
 for demonstrating autonomous vehicle safety.</code></pre>

&gt; That&#x27;s a very bad look; the whole point of self-driving cars is that they can react to unexpected circumstances much more quickly than a human operator, such as when someone walks out into the road.No, the whole point of self-driving vehicles is that firms operating them can pay the cheaper capital and maintenance costs of the self-driving system rather than the labor cost of a driver.

FTA:&quot;The Uber vehicle was reportedly driving early Monday when a woman walking outside of the crosswalk was struck....Tempe Police says the vehicle was in autonomous mode at the time of the crash and a vehicle operator was also behind the wheel.&quot;That&#x27;s a very bad look; the whole point of self-driving cars is that they can react to unexpected circumstances much more quickly than a human operator, such as when someone walks out into the road. Sounds like Uber&#x27;s platform may not be up to that standard yet, which makes me wonder why they&#x27;re on public roads.On the other hand, it sounds like it happened very recently; I guess we&#x27;ll have to wait and see what happened.

On the bright side, with an autonomous vehicle, an accident is a learning experience that will benefit other autonomous vehicles.

Can you provide a source? From all the articles I can find, the language is all pretty much the same:&quot;Tesla driver killed while using autopilot&quot;&quot;The Tesla driver killed in the first known fatal crash involving a self-driving car&quot;&quot;It&#x27;s been nearly a year and a half since Joshua Brown became the first person to die in a car driving itself.&quot;&quot;Tesla bears some blame for self-driving crash death, feds say&quot;

Tesla was not a self driving car, please stop confusing this.This case is in fact the first autonomous test vehicle caused fatality.There were at least several Tesla autopilot related fatalities and injuries, but I would seriously not put those into self driving bag.

Interesting, apparently there&#x27;s no way to know if autopilot was enabled or not.&gt; A Tesla spokesperson said the car in the Chinese wreck was badly damaged and unable to transmit data. “We therefore have no way of knowing whether or not Autopilot was engaged at the time of the crash,” the spokesperson said in a statement.

There may be more than that. This one wasn&#x27;t as heavily covered <a href="https:&#x2F;&#x2F;www.mercurynews.com&#x2F;2016&#x2F;09&#x2F;14&#x2F;second-fatal-crash-may-involve-tesla-autopilot&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.mercurynews.com&#x2F;2016&#x2F;09&#x2F;14&#x2F;second-fatal-crash-ma...</a>

Tesla Autopilot != Self-driving(full autonomy)

The second ever [1] fatal car crash involving a self-driving car. This is the first one where a third party is killed.[1] <a href="https:&#x2F;&#x2F;www.theguardian.com&#x2F;technology&#x2F;2016&#x2F;jul&#x2F;01&#x2F;tesla-driver-killed-autopilot-self-driving-car-harry-potter" rel="nofollow">https:&#x2F;&#x2F;www.theguardian.com&#x2F;technology&#x2F;2016&#x2F;jul&#x2F;01&#x2F;tesla-dri...</a>

Again... &#x27;Uber has temporarily suspended the testing of its self-driving cars following the crash of one of them in Tempe, Ariz. The ride-hailing company had also been conducting experiments in Pittsburgh and San Francisco.&#x27;I was resumed shortly after.27 Mar 2017 (1 year ago)<a href="https:&#x2F;&#x2F;spectrum.ieee.org&#x2F;cars-that-think&#x2F;transportation&#x2F;self-driving&#x2F;uber-crash-leads-company-to-suspend-robotaxi-testing" rel="nofollow">https:&#x2F;&#x2F;spectrum.ieee.org&#x2F;cars-that-think&#x2F;transportation&#x2F;sel...</a>

that&#x27;s SOP for every av program. have incident? ground the fleet. doesn&#x27;t matter why, or who&#x27;s at fault.

Important and and missing:&quot;[Uber] said it had suspended testing of its self-driving cars in Tempe, Pittsburgh, San Francisco and Toronto&quot;[1]1: <a href="https:&#x2F;&#x2F;www.nytimes.com&#x2F;2018&#x2F;03&#x2F;19&#x2F;technology&#x2F;uber-driverless-fatality.html" rel="nofollow">https:&#x2F;&#x2F;www.nytimes.com&#x2F;2018&#x2F;03&#x2F;19&#x2F;technology&#x2F;uber-driverles...</a>

Perhaps they were walking their bicycle across the street.

Seems very unlikely to me that this case will be brushed off without police looking at video.

It would explain the &quot;outside a crosswalk&quot; remark better if the person were riding the bike. This is one of those cases where video evidence should be subpoenaed, though. It&#x27;s far too common for the police to accept the statement from the only party still alive, and come up with incorrect conclusions about what happened on the street.

A police officer in Nashville shouted at my wife for riding her bike in the road in accordance with Nashville law. Basically, he didn&#x27;t know the law he was enforcing.

Unfortunately, many police forces don&#x27;t train the police on how to handle bicycles. For example, in many places riding on the sidewalk is illegal, yet police will suggest a cyclist get off the road and ride on the sidewalk.From studying many bicycle-related police reports, I was dismayed at how often I read the phrase &quot;a pedestrian riding a bicycle.&quot;

That&#x27;s dumb. The term means someone who is walking. &quot;Per pedes&quot; means &quot;by foot&quot;.

When I lived in California the press reported pedestrian and bicyclist deaths under the same &quot;pedestrian&quot; moniker.

I&#x27;ve heard that police reports often refer to all non-drivers as &quot;pedestrians&quot;, even if they were riding a bike (which I learned after a neighbor was killed while riding a bike).  I don&#x27;t know if this is legalese or police jargon, but either way you would hope a journalist who covers these things could translate.

The video says bicyclist and shows a bent bicycle. So pedestrian or bicyclist?

Given that so many of autonomous car accidents logged by the California DMV seem to be the fault of the human driver trying to take control, I’m surprised that the accident here reportedly happened in autonomous mode, which for all its flaws seems to have been good at not hitting things in front of it. In fact, seems like most of the CA accidents were the autonomous car being rear ended or side hit from other cars, usually for going too slow or stopping unexpectedly.

Given how incompetent Uber is at regulation of anything else in their company, is this really a surprise?

It&#x27;s not about sympathy. It&#x27;s about one question: do they kill fewer or more people than humans per mile?If they kill fewer people, they can be run by a joint venture of Satan and the Mafia for all I care.

Having a hard time sympathizing here. It&#x27;s one thing to fear air travel based on the very few but very publicized plane incidents (considering all the data we have on the safety of air travel). It&#x27;s another thing to hold these self-driving car companies accountable, considering a) the lack of data and history of such programs, and b) their touted benefit as a safer alternative to human-driven cars.

As they should: humans fail independently, but autonomous cars can fail systemically, including hacking.  Bicycles are known to be tricky for autonomous cars.

Wow. This is dark but it feels real to me.I had been assuming that self driving cars would be a &#x27;reset&#x27; for our decision to value cars more than the countless lives they end. Self driving cars could take the irrationality and emotion out of driving. It would be America&#x27;s &quot;stop the murder of children&quot; moment (The Netherlands rallying cry to reform auto traffic in 1970&#x27;s).But now I see that it can, and probably will go the other way. These giant companies could push to further restrict the rights of cyclists&#x2F;pedestrians around roadways, vilify the &#x27;stupid&#x27; people who made a reckless decision that resulted in an accident, and push our country even further into car culture.

Even if its provable that self driving cars are 10x safer, every accident of this unfortunate kind will be used against autonomous cars. I don&#x27;t see how to overcome this -- I don&#x27;t expect reason will overcome the perceived danger.

&gt; You&#x27;re goddamn blind.We ban accounts that do this. Please read <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;newsguidelines.html" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;newsguidelines.html</a> and follow the rules when posting if you want to keep commenting here.

You&#x27;re goddamn blind. Yes, of course it matters more if I&#x27;m killed by an AI machine than it does if I&#x27;m killed by another human being. This is a new development, humans killing humans are not. What if it was autonomous warfare? It is surely coming about in a very near shipment, SOON. The world and history are not static, and techno-utopia is big fat lie, and the idea that things just get better as time advances frontwards is plain infantile. Reality is way more fucked up than you can imagine.

It&#x27;s a sad reality, but regardless of how many times per day a pedestrian is hit by a human-driven car, such incidents involving self-driving cars will be headline news for years to come.

There aren&#x27;t a lot of options when you need to have cars turning on roads with bike lanes. At some point the car lane and the bike lane have to cross paths unless you&#x27;re going to build a fully grade-separated bike lane.I suppose you could have a wall that separates the bike lane from the road except for a short area where cars can cross over, so that at least there&#x27;s only a small space where the bikes &amp; cars can interact, but that introduces a bunch of new problems (cars that can&#x27;t see bikes behind the wall when they go to cross over, cars that run into the wall, probably even more swerving to get into the lane because your choices are do it RIGHT NOW or you&#x27;re stuck&#x2F;hit a wall).

Yes, it&#x27;s been proven that un-protected bike lanes such as those are very dangerous. There are more and more separators coming out (<a href="https:&#x2F;&#x2F;peopleforbikes.org&#x2F;blog&#x2F;a-new-generation-of-bike-lane-separators-is-starting-to-arrive&#x2F;" rel="nofollow">https:&#x2F;&#x2F;peopleforbikes.org&#x2F;blog&#x2F;a-new-generation-of-bike-lan...</a>) to allow for protected bike lanes.However, I do need to mention that when you choose to bike on the sidewalk, you&#x27;re creating the same type of problem that cars that ignore traffic rules create for bikes. There&#x27;s a reason why biking on the sidewalk is illegal in most places, and why bike lanes&#x2F;roads exist.

I don&#x27;t think that&#x27;s insensitive to the death. It&#x27;s properly placing (partial) blame on the cause of the accident.A better design for that intersection would be to keep the bike lane against the sidewalk and the turn lane inside of that, so that they don&#x27;t cross, and have a &quot;no turn on red&quot; sign, so that the paths of bicyclists and people turning in cars never cross as long as everyone obeys the stoplight and sign.

AZ driving rules language on solid white lines is the same as FL (where I live). They use the words &quot;should not cross&quot;, so it&#x27;s not technically an infraction for a driver to cross a solid white line. I was in an accident once due to someone crossing a solid white while at a light and cutting me off so I&#x27;m familiar with the RAW on those lines.

I&#x27;m much more a fan of the design that places a buffer (either sidewalk or shrubs) between the CARS and any other type of vehicles.  Sidewalks are tolerable as they are often raised above street grade and a vehicle usually has to be actively controlled to overtake such a raised divider.

I have been hit by a car in a bike lane by a car making a rogue left turn (In the UK).  I&#x27;m always wary of riding in bike lanes like this now.

Those sort of bike lanes teach cyclists that they should under take slower moving traffic  - the problem is the majority of deaths in London are due to vehicles turning into undertaking cyclists.

Anec-dataly, I agree.  We live at the edge of a residential area where there is a long stretch of road with bike lanes and turns into the residential area every block.  Not only do people regularly drive in the bike lanes, the people who do so are typically the most aggressive and inattentive drivers who are trying to angrily speed around the lines of cars who are driving the speed limit and respecting the stop signs.  I&#x27;m not aware of any recent accidents, but only because cyclists seem to just completely avoid the entire stretch of road.

That looks completely identical to all bike lanes I&#x27;ve seen in suburban areas. I think bike lanes are just inherently dangerous. Would be interesting to see what a well-designed suburban bike lane looks like.

Not to be insensitive to the death. But, that bike lane design is just horrible. Are human drivers really known for respecting the &quot;do not cross solid white line&quot; rules? 9&#x2F;10 drivers that realize at the last moment they need to turn right are going to quickly cross over this bike lane probably without much time to check their blind spot. If I was cycling on this path, I&#x27;d want to use the sidewalk and be extra cautious of peds, I don&#x27;t trust drivers human or not.

This is both terrifying and normal. The bike lanes in my area are frequently used as turn lanes... by vehicles that don&#x27;t fit in them - despite being clearly marked as bike lanes.Human drivers are terrible.

The car turn-lane is to the right of it, so at least it&#x27;s maintained as part of the road.  Bike lanes that are truly a shoulder with bike symbols painted on are often unmaintained, and full of pot-holes, debris, and such.

Yeah, they do the similar stupid &quot;bike lane&quot; crap here in Indiana as well.Yeah, like a line is going to protect me from run amok car drivers who are too buy on their cell phones. Or it&#x27;s going to stop the bus drivers from using them as pickup&#x2F;dropoff areas, cutting bicyclists off.Long story short, I bike on the sidewalks. I can be a responsible biker on sidewalks while heeding to pedestrians. My life is more important than to be biking on a vehicle road.

I&#x27;ve actually never seen a bike lane that looks different than that...

How about a &quot;sharrow&quot;, where cyclists drive right-of-center, in hopes of avoiding the &quot;parked car door lane&quot;, and encouraging drivers to cross the double-yellow to overtake.<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Shared_lane_marking" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Shared_lane_marking</a>

That bike lane is just a shoulder with bicycle symbols painted in it and it looks terrifying.

Yes, I bike almost daily on Venice Blvd in LA, and every intersection I encounter is setup like the one discussed here, along with buses weaving in and out. A vicious battle flared up last year when the city replaced a lane of traffic on Venice with protected bike lanes (some discussion here <a href="https:&#x2F;&#x2F;la.curbed.com&#x2F;2017&#x2F;10&#x2F;25&#x2F;16528864&#x2F;road-diets-los-angeles-vision-zero" rel="nofollow">https:&#x2F;&#x2F;la.curbed.com&#x2F;2017&#x2F;10&#x2F;25&#x2F;16528864&#x2F;road-diets-los-ang...</a>). There seems to be a marked increase in the hostility between cyclists and drivers now.

This design seems pretty common in the US.  Some bike lanes on major streets in LA are like this, except there they also have to contend with buses pulling into stops.  Biking on those streets would be terrifying, but I saw people doing it.

Unfortunately, this is a design that is pretty common in many of Tempe&#x27;s major intersections. I pass through a similiar intersection biking home from ASU and have had multiple close calls with human drivers.Anecdata: The times I have interacted with one of Uber&#x27;s autonomous vehicles at this intersection, they tend to hit the brakes pretty hard as they are about to enter the turn lane and I am around 10ft behind them. All of my interactions with them, however, have been during the day, and it was unclear if it was the human driver in control.

The video says bicyclist and shows a bicycle, while the written article says pedestrian and said the car hit &quot;a woman walking&quot;... I&#x27;m confused as to why there is a contradiction, but considering the video shot of the bicycle I assume you&#x27;re correct

The video mentions she was on a bike and shows a mangled bicycle. Also the line markings on the street indicate there is a bike lane to the left of a right turn lane. Definitely very bad.<a href="https:&#x2F;&#x2F;www.google.com&#x2F;maps&#x2F;@33.4369934,-111.9429875,3a,75y,83.99h,72.74t&#x2F;data=!3m6!1e1!3m4!1scUyILaxFs5z63AL2SupCJw!2e0!7i13312!8i6656" rel="nofollow">https:&#x2F;&#x2F;www.google.com&#x2F;maps&#x2F;@33.4369934,-111.9429875,3a,75y,...</a>

It seems like a bunch of what-ifs that normally come up with self-driving cars are about to get answered and precedents are about to be set.I assume this case will also be one of the most well-recorded cases of a fatal car accident in history as well, given the amount of sensors and equiment on-board a self-driving car, along with eye witness testimony from the operator on board.Can&#x27;t tell if Uber has just been incredibly unlucky as of late or if just enough of their employee-base is incompetent as to prevent them from just having a quiet year with no large failures.

Maybe deaths per speed per time (or distance, since speed over time is the same as distance at speed). More deaths at higher speeds and fewer at lower would seem more likely. Purely using time or distance may skew the interpretation, if there are more deaths at some speed ranges than others.

Perhaps time as opposed to distance?  After all, we drive much fewer miles in a city, but there are much greater opportunities for accidents, particularly pedestrian accidents.

Yes, these statistics are typically based on miles driven.

It occurs to me that because the deaths caused by autonomous vehicles may not follow the same distribution across types of deaths, it might make more sense to compare total deaths per million miles between human and autonomous drivers.

Wow.If deaths are broken out by pedestrian vs passenger, I wonder how people will respond if safety skews heavily towards one of those groups with self driving cars.

This is an interesting metric. Although self driving test cars are rare, the whole point of their existence is to drive so they probably clock more hours than a normal car.About 5400 pedestrians are killed each year in the US.
US drivers go 3.1 trillion miles a year. So they kill a pedestrian about every 5.7 billion miles. Last November, Waymo said they had 4 million self driven miles, so well short of statistically expecting to hit a pedestrian. In September of some unspecified year Axios claims Uber had self driven over 1 million miles.My estimate needs help, the distribution of pedestrian bearing roads and pedestrian free roads likely does not match from my total miles per year number and what carbot testers cover. Also, this may have been a cyclist death, which adds another 800 or so deaths per year.But, in any event, in rough numbers, Uber appears to have beat their expected time to pedestrian fatality by two or three orders of magnitude.

Distance is the metric used so far but also isn&#x27;t appropriate for comparing self-driving cars to human driven cars. The human driven bucket contains all miles driven, highway cruise control, light snow, heavy rain, tricky merges, etc. The self-driving metric is only the easiest possible miles. Overtime those miles will expand and harder scenarios will be incorporated but to really know if self-driving cars are safer we need apples to apples comparisons which is going to require matching humans vs robots on miles driven and a categorization of those miles, maybe a count of tricky unexpected scenarios as well.

Curious how the numbers work out of number of miles [1] to number of deaths of autonomous vs standard vehicles.I know the scale is different but since this is the first death I’m curious if the percents fall in line.[1] is distance the right metric?

You mentioned bad luck and that could very well be it. Unless Uber starts having more incidents, it isn&#x27;t fair to say that Uber&#x27;s technology is inferior.

Why am I not surprised that, among all the teams competing in this space, many of which have clocked in far more miles than them, Uber is the first one to kill someone?This may well just be bad luck. But I cannot shake the feeling that if Uber started an ice cream venture, they would store their molasses on a hill in a Boston. The only way to get “humanity” associated with Uber would involve an Uber zeppelin.Or it’s a conspiracy, because nothing is as threatening to Uber than autonomous cars. This is certain to invite more regulatory scrutiny. Just kidding... I think..

Calling these pre-programmed multi-MJ wheeled computers &quot;self driving&quot; is foolish and deceptive.

&gt; if this is the first (known) case of A.I. controlled cars killing humans, the number of casualties can only grow from now onwardsThis is how counting works.

Every year between 30000 and 40000 Americans are killed in car accidents. With more self-driving cars on the road it comes as no surprise that self-driving cars will cause some of them.

Not to be the guy who says &quot;I TOLD YOU SO&quot; but it had to happen at some point, sadly &amp; unfortunately - if this is the first (known) case of A.I. controlled cars killing humans, the number of casualties can only grow from now onwards

Thank you for that link. I will pass it along to my former colleagues who I suspect will be very interested in her work.

Dr. Mary Cummings has been working on talking to NHTSA about implementing V and V for autopilots&#x2F;AI in unmanned vehicles for a few years now. She&#x27;s also been compiling a dataset exactly like what you are talking about.I think the idea is to build a &quot;Traincar of Reasonability&quot; to test future autonomous vehicles with.You might want to check out her research <a href="https:&#x2F;&#x2F;hal.pratt.duke.edu&#x2F;research" rel="nofollow">https:&#x2F;&#x2F;hal.pratt.duke.edu&#x2F;research</a>

You ultimately have to at some stage, since any test track is a biased test by its nature.It is more an issue of how sophisticated these vehicles should be before they&#x27;re let loose on public roads. At some stage they have to be allowed onto public roads or they&#x27;d literally never make it into production.

How about not testing it on the unpaid public in incremental patches like this age of software &quot;engineering&quot; has decided it was a good idea to do?

It would be required by law. Violating the law would hold corporate officers criminally responsible.

I guess my hopeful answer would be they start getting treated like Enron and someone high up goes to prison until they start to comply.

Why would Uber agree to such regulations?They were unwilling to legally obtain a self-driving license in California because they did not want to report &quot;disengagements&quot; (situations in which a human driver has to intervene).Uber would just set their self-driving cars free and eat whatever fine&#x2F;punishment comes with it.

To ensure that all automotive software incorporates lessons learned from such fatalities, it would be beneficial to develop a common data set of (mostly synthetic) data replicating accident and &#x27;near miss&#x27; scenarios.As we understand more about the risks associated with autonomous driving, we should expand and enrich this data-set, and to ensure public safety, testing against such a dataset should be part of NHTSA &#x2F; Euro NCAP testing.I.e. NHTSA and Euro NCAP should start getting into the business of software testing.

Self-driving Uber car kills Arizona woman crossing street