Will this article make you think twice about autopilots and other ubiquitous types of automation?

It is cause for concern, certainly.
I am a www refugee of the banking system collapse when I went through 3 successive mergers from World Savings that merged into Wachovia then into Wells Fargo. Between the World and Wachovia merger I completely disappeared from the system for about 72 hours and was in a state of complete panic as all of my savings accounts, CDs etc. went poof.
Of course I had a paper trail but still….the system could not “account” for 130k for some days!

Of course I have still lost my bra in this economy, but at least I KNOW what happened to my money. lol

I’d like to offer a dissenting opinion regarding the autopilot example. I go to school and work with a lot of pilots, have spent some time in the cockpit myself, and I can tell you that the example the author provides isn’t a matter of the autopilot fucking up, it is a matter of the pilots lacking situational awareness and making the wrong decision; as the article so explicitly states itself.

What I find confusing about the article though is that the examples it gives of autopilot failure are all result in pilot error and disaster, and then it advocates removal (or being wary) of the autopilot, exponentially increasing pilot workload and the possibility of pilot error. Something like 80% of aviation accidents happen when pilot workload is the highest (takeoff and landing), and even though regional pilots (like the first example in the article) need to have roughly 1000 hours before getting hired by a regional, they are still prone to mistakes. Even though autopilots throw another system to fail into the equation, looking at what’s been presented in FAA Reports (warning, that’ll download a PDF) in the past, I think you’d be hardpressed to find any high ranking official at Boeing, Airbus, the FAA, or an airline who advocates removal of the autopilot because it’s “safer”.

@DeanV, Of course you’re entitled to your opinion but I feel you may have missed the overall point of the article, which is incapsulated in its title.

@Pachyderm_In_The_Room I’m certain I did. My point was that they seem to be sort of off base on the whole autopilot thing, which weakens the article IMO.

There is zero reason to think that just taking away the autopilot would improve any of the situations described.
Better training and more emergency-situation training in simulators is clearly what’s needed. And if they’re going to improve training then there’s every reason to think that a better-trained pilot, along with an autopilot, is the best preparedness you can achieve.

I think @DeanV is onto something. The core point of the article is not that there’s something wrong an autopilot or any other automatic system, it’s that people forget how to do stuff when they don’t have to anymore. That’s true.

But it’s a mixed bag at worst, automation frees us up for other activities for which we’d not have bandwidth if we’re dealing with the mundane.

Most of us who drive cars would not be using one if we had to deal with as much continuous engine monitoring and adjustment as an early driver-chauffeur used to. Not knowing how the car works is occasionally a fatal problem but far more often the automatic systems enable us to use the car when otherwise we’d not have a clue.

Nothing is perfect. I heard of an automatic steel cutting machine that would follow a pencil line with an electronic eye for precise cuts. Worked GREAT! Until a fly landed on the steel and crossed the electronic eye. (And everywhere the fly went the torch was sure to follow.) An AI system would help a lot if the system could also reason. Then any flies would/could be ignored.

I saw that article the other day. We’re all victims – some more than others – of this phenomenon.

How many people still remember how to read a map? – if you can find one in the first place. GPS are great, but supposed your battery dies? How many people trust their calendars to Google or Outlook or (god forbid) their smartphones? What about address books – suppose your phone dies? How would you know how to reach someone?

In the airplane examples in that article, one has to be extremely critical of the airline companies. Electronics are supposed to be in addition to – not a replacement – to standard procedures and common sense.

I feel that computer errors get a lot of press, but human errors happen all the time under media radar. I work in an er, I see all the local car accidents. There are a lot. None of them are newsworthy. But the first car accident the Google auto driving car makes, stop the presses. But I still feel its safer. We just are very reticent to give up control.

Good point @drhat77 there will be big attention on the first Google car accident(s) which will inevitably happen when that technology spreads.
There may be thousands of lives saved yearly where the auto-car is driving better than regular folks, but people will go nuts especially when the auto-car makes a mistake that people wouldn’t have.

Deleted by Pachy

@DeanV, I’m not a pilot and thank God have never been in a crash so I won’t debate you about autopilots, but there are other examples in the article I’ve experienced. Too many times, for instance, trusting a spell checker and winding up with a word in the sentence that totally changes its meaning – and me, an English major and onetime spelling bee champ who has been professionally writing for 50 years!). And too many times, relying on the GPS on my iPhone or in my car and ending up at the wrong address.

Far more seriously, a year ago I got x-rays for an unusual pain in my legs and was told by the doctor the pictures revealed no problem … and then a second doctor the next day surmised I had a malady that wasn’t showing up and ordered me immediately to the hospital, where I was treated for three days. I’m now on a lifetime med for that.

I’ve had my driving license for over 50 years, yet other than distantly spaced renewals and periodic traffic safety classes, neither of which do not require me to demonstrate actual road skills, only vision ability and memory (both of which I’ve come to trust less and less), I worry that my driving ability is nowhere near as sharp these days as it has to be, especially since, unlike in my younger years, I see drivers all around me talking on cell phones, texting, and staring at GPS displays (I almost got hit by one such “driver” yesterday). I won’t even get into the discussion of how much I fear the inevitable availability of self-driving cars.

In the article the author says, “Psychologists have found that when we work with computers [which are the brains for most of the technologies cited in the article], we often fall victim two cognitive ailments – complacency and bias – that can undercut our performance and lead to mistakes. Automation complacency occurs when a computer lulls us into a false sense of security. Confident that the machine [and software] will work flawlessly and handle any problem that crops up, we allow our attention to drift (we’ve all done that in our cars; just yesterday, I found myself one exit past my own because I had been listening intently to an interesting NPR story]. We become disengaged from our work, and our awareness of what’s going on fades. Automation bias occurs when we place too much faith in the accuracy of the information coming through our monitors [on n x-ray, for example]. Our trust in the software becomes so strong that we ignore or discount other information source, including our own eyes and ears [and I would add experience].

Sorry to go on and on, but I feel very strongly about this topic. Guess we’ll have to agree to disagree, but that’s the great thing about Fluther, isn’t it?