[tokenadult]

Here are some facts and figures about driving versus flying for travel safety per passenger mile over a time period carefully measured by official statistics. "In absolute numbers, driving is more dangerous, with more than 5 million accidents compared to 20 accidents in flying. A more direct comparison per 100 million miles pits driving's 1.27 fatalities and 80 injuries against flying's lack of deaths and almost no injuries, which again shows air travel to be safer."[1] I thought everybody knew that airline travel is by far the safest form of travel.[2]

The basic issue about self-driving cars is that setting the standard of performance at "better than most human drivers most of the time" is a moderately low technical bar to clear, but would still result in a HUGE reduction of deaths, injuries, and damage to vehicles and roadside property. But I'm still not sure how soon self-driving cars will be good enough to drive on snowy roads (Google's self-driving cars have never been tested in snowy conditions, as of the last time I checked) or even in rain.[3] The story posted today is more nuanced than a story posted yesterday about Elon Musk's predictions, and reports, "He said he believed that we're still 20 years out from the roads being full of autonomous cars."

[1] http://traveltips.usatoday.com/air-travel-safer-car-travel-1...

[2] http://www.businessinsider.com/flying-is-still-the-safest-wa...

[3] http://spectrum.ieee.org/cars-that-think/transportation/adva...

[jowiar]

Regarding poor-weather performance, I'm assuming self-driving cars will outperform human drivers sooner than in good conditions, rather than later (as long as it is made a priority). The information a self-driving car can gather and process about wheel traction is far more detailed than can be gathered by human drivers.

[Alupis]

We're largely talking about only the US market here -- in which only (and I say that word relatively) 32,000 people died in any sort of automobile related accident last year.

To put this in perspective, the common Flu (a totally vaccinatable, preventable, and treatable illness) killed more than 52,000 people in the US last year alone.

In reality, automobiles in the US are the safest they have ever been since 1949, and automobile related deaths have dropped more than 26% since 2005 alone. According to the CDC, automobile related deaths are not even in the top 10 leading causes of death in the US.

Last year, heart disease was responsible for 596,577 US deaths last year alone. Stroke claimed 128,932 US lives. Diabetes took 73,831 american lives. Suicide robbed 39,518 people their life last year -- in excess of 7,000 more than died by automobile related accidents.

In 2009, approximately a grand total of 2,436,652 died in the US of all causes (including natural causes). Of that number, automobile related deaths only accounts for 1%. Of the entire US population, we're talking about 0.0096%...

According to the CDC, in 2012, 10,322 people (or 31%) of all automobile related deaths were alcohol-related deaths. Another 18% of automobile related deaths are attributed to drugs other than alcohol in the same year. That accounts for just about 50% of all vehicle related deaths, leaving the true automobile related deaths due to accident at a shockingly low number (relatively) of about 15,000 people per year.

All of this is to say, relatively speaking, automobile related deaths are not a significant concern. It does not make sense to spend significant effort in this sector while ignoring or not increasing efforts in these other much more problematic areas.

Sources:

[1] http://www.cdc.gov/nchs/fastats/leading-causes-of-death.htm

[2] http://www.usatoday.com/story/news/nation/2013/01/09/guns-tr...

[3] http://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_in...

[4] http://www.businessinsider.com/top-causes-of-death-united-st...

[5] http://www.iihs.org/iihs/topics/t/general-statistics/fatalit...

[6] http://www.cdc.gov/Motorvehiclesafety/impaired_driving/impai...

[apendleton]

Automobile-related deaths are the number one killer of healthy adults in the US, and the number one killer, period, of Americans between the ages of 5 and 34. (sources: http://www.ted.com/talks/jeff_speck_the_walkable_city and http://www.cdc.gov/injury/overview/data.html respectively). They're not insignificant, especially when you look at their value in terms of dollars per life saved for prevention, or (even moreso) years of life lost per death, as automobile deaths disproportionately affect young people, as compared to, e.g., stroke.

Also, I have no idea why you exclude alcohol- and drug-related deaths from "true" automobile deaths. Humans are fallible. They get tired, get distracted, see poorly at night, misjudge speed, and make stupid decisions like texting while driving, and, yes, drinking and driving. Mitigating that fallibility is the entire value proposition of the self-driving car. Deaths due to drunk driving are absolutely among those that self-driving car advocates aim to prevent.

[Alupis]

> Deaths due to drunk driving are absolutely among those that self-driving car advocates aim to prevent.

Instead of proposing legislation that everyone must own a self-driving vehicle (the path self-driving vehicle advocates seem to be heading towards), wouldn't a more reasonable, less expensive, and less obtrusive solution be to mandate breathalyzers in every vehicle? (I don't support this either, but it's less expensive and less of a burden).

Self-driving cars won't last as long due to the sheer amount of technology required to be in the vehicle. So instead of getting 5-20 years of out of your "dumb car" you'll get closer to 2-5 years before the electronics inside degrade to an unsafe state.

Can the electronics be upgraded/replaced when they become unsafe? Sure, but this is a significant cost burden on the consumer... The precision and reliability the electronics must provide has to be absolute. To achieve this level, they'll have to be better than military-grade sensors -- heck, military helicopters with some of the best sensors, ai, and electronics crash routinely... and they get regular maintenance before and after every flight. How long do we really expect the average consumer's self-driving vehicle to last when most drivers don't even change their oil once every year?

> Automobile-related deaths are the number one killer of healthy adults in the US, and the number one killer, period,

That's a disingenuous statement. People were healthy until they contracted the flu. They were healthy until they suffered a sudden heart attack. Those are the true leading killers of all Americans... healthy and not.

[apendleton]

> Instead of proposing legislation that everyone must own a self-driving vehicle (the path self-driving vehicle advocates seem to be heading towards)

I think the path many advocates are headed towards is actually more radical: largely eliminating personal vehicle ownership entirely, with rides available as a service, Uber-like, as part of most people's multimodal transportation mix that might also include transit, more walking, etc. I don't expect either direction to occur via a mandate, though, but the economics of the more aggressive plan are interesting. Most cars sit unused for 22 hours a day, and there are enormous economies to be gained from improved utilization if cars are a shareable asset, especially with the labor taken out of the equation (as compared to current Uber), and I think the shift may well happen on its own, at least in part and in some places. And to be clear, it's not an all-or-nothing proposition; replacing some drivers still potentially improves everyone's safety.

As for the second bit: if you think my first stat is disingenuous, pay attention to my second, which you cut off halfway through. In that age bracket, automobiles are the top killer overall, including health-related causes. The flu disproportionately kills infants and the elderly (and for what it's worth, you vastly overstated the efficacy of the flu vaccine in your original post; it's totally worth getting, but still only decreased recipients' chances of contracting the flu by a bit more than half; "totally vaccinatable [and] preventable" it is not).