I do as well, especially with the run roof open and windows down 
…
1 Like
Nope. My driving mood depends on various factors. A robot cannot get to my mood unless I tell it. It’s not a pattern that it can be predicted.
It’s not about mood. It’s about the computer doing things like analyzing your eye movement, head movement, and driving behavior as you’re driving. If it sees enough things it doesn’t like, it can apparently shut things down.
It will be interesting to see how a computer will distinguish between someone under the influence of drugs and someone in distress that is trying to get to a hospital.
From some other articles I’ve read, it’s an active question as to whether it’s even desirable to do so. Any unsafe driving activity is considered a problem, and the fact that it casts a wider net than “people on drugs” is a claimed feature - not a bug.
Many years ago one of my teachers drove himself to the hospital after being severely wounded in the neck by a chainsaw. I’m not sure I would be in favor of a system that would have condemned him to death because he had to drive through a locked gate, etc. to get help.
But enough of what I think of this.
1 Like
It would take exactly once that the computer decided it didn’t like my driving before I disabled that permanently.
1 Like
My opinion exactly. Although, again, this isn’t for the benefit of the person driving the car - so the core idea is for manufacturers to design it such that it can’t be disabled.
Nice! Three on the tree. Very cool. 
1 Like
What’s more cool is 4 on the tree. Old Saab’s had a 4 speed with a column shift. Friend in high school had one that I drove. And yes, it was 50 years ago.
1 Like
Nice!
What’s less cool, is driving home from highschool in my bug, the stickshift came out of its socket, leaving me in first gear with the stickshift in my hand! 
A slowwwww drive on the side of the road all the way home.
But also my first credit union loan for $300 to get it fixed at a long-gone Volkswagen dealer in town
@webwalrus:
And no, I don’t think the tech is ready, nor am I a fan of a car making decisions for me anyways.
For cargo transport, perhaps. Would I want autonomous semis on the road? I don’t think I’d trust that much. I’m less put off by the idea of smaller unmanned transport like box trucks or step vans.
Bless your heart 
I think that you’re brining up a really, really important point here. The example you gave is very personal and meanignful to you, but also very much an outlier when considering the harm that’s cause by imparied driving (especially when you include impairment due to fatigue). How could you possibly be in favour of something that you know would have resulted in the death of someone close?
Another way of looking at the same question is to ask how acceptable people would find it to trade 25,000 lives (significanlty less than the anual automotive death toll in the US) randomly chosen in a lottery for perfect automotive safety for a year for the rest of the population. I suspect that, despite the benefit of fewer deaths, most of us find the idea abhorent. I do.
When making decicions about risk, most people have a bias to purchase agency with added risk. As technology will very soon get to the point that activities like driving (or in this case, the determination as to whether a person is fit to drive) will become objectively less risky by having them handled by autonomous systems, topics like this become more important to discuss and think about.
How much agency (free will, in some sense) are we willing to trade for safety? How much should we? What are the unforseen societal consequences for doing so? I have no idea what the answers are, but they’re really interesting questions and incredibly important ones to consider. And, until they’re answered, I don’t think that we’ve had enough of what you think about this
1 Like
Actually, stuff like this “attention detection” software would probably be more appropriate to roll out in semis and commercial vehicles to start with. There are far fewer edge cases around commercial transport than there are for passenger vehicles.
And we all know programming is about 10% figuring out what “typical” is, and about 90% figuring out all the weird ways that “typical” doesn’t apply - and coding defensively. 
Somebody with a life threatening injury that wants to drive (distracted) to the hospital is very, very unlikely to be jumping into a semi. And a parent with a fussy toddler that’s causing their attention to drift probably isn’t going to be taking them to preschool while hauling a huge truck full of cargo.
1 Like
Agreed on those counts, but the thought of a 26,000+ pound vehicle driving itself gives me pause.
Didn’t you see Maximum Overdrive??!?
The other issue is that even if we made these safety features available tomorrow on every new car purchased, I’d wager it still doesn’t prevent most of the drunk driving until those cars filter down to the rest of the population and cars without the feature cease to be generally available.
This means that at the beginning, it’s almost all downside. Even though we’d probably still find it abhorrent, we wouldn’t even be able to make your “25,000 lives for perfect safety” choice for at least a couple more decades, even if the tech was rolled out tomorrow.
The other thing this really doesn’t take into account is relative likelihoods of injury to others. And that’s pretty important to the value calculus, IMHO.
For example, somebody driving home drunk from a bar in New York is far, far more likely to be in a position to cause harm to others than somebody who had a few beers (and is thus probably not in great shape to drive) but suddenly needs to get a family member to the hospital in, say, rural Wisconsin.
If you asked me, I’d tell the New York person to call a cab and tell the person in rural Wisconsin to drive their family member to the hospital - possibly after calling ahead.
Exactly. If we’re at the point where the car can say “hey, it looks like you’re not doing so good - let me drive. Do you want to go to the hospital?”, and if the car will get them there safely and reliably, then the calculus changes massively.
But right now, some of the assumptions that people make - like “why not just call an ambulance or a taxi?” just don’t practically apply in large swaths of the United States.
1 Like
Autonomous cargo carriers may be coming sooner than you think. There is a strong enough economic motive behind autonomous cargo that probably makes it inevitable.
Among nearly a dozen companies developing autonomous trucking, San Diego–based TuSimple is trying to get ahead by combining unique technology with a series of strategic partnerships. Working with truck manufacturer Navistar as well as shipping giant UPS, TuSimple is already conducting test operations in Arizona and Texas, including depot-to-depot autonomous runs. These are being run under what’s known as “supervised autonomy," in which somebody rides in the cab and is ready to take the wheel if needed. Sometime in 2021, the startup plans to begin doing away with human supervision, letting the trucks drive themselves from pickup to delivery without anybody on board.
This Year, Autonomous Trucks Will Take to the Road With No One on Board - IEEE Spectrum
Then it’s clear to me, those people haven’t seen the movie either 
I am not an experienced truck driver, but for a time I had a commercial license and drove trucks with 40’ trailers. IMO, autonomous trucks should be able to operate safely on interstate (I.e limited access) highways long before cars will be operating on crowded streets. Especially those that are remotely monitored by experienced truck drivers who could take over remotely if needed.
Are you mostly seeing things like intersections and traffic lights as the challenges for the autonomous systems?
I live by a major city, and the interstate highway system here is used by the average person for day-to-day commuting. It’s definitely doesn’t seem less crowded - especially at “peak” times of the day. There are days when the interstates are, quite literally, a parking lot.
Was it a Ford Consul by any chance?