[EazyC]

Are you sure things like this can't work? I work in a computational neuroscience lab and we do a lot of modelling and also machine learning. Things like Eulerian Video magnification (EVM) already prove you can do the rudimentary proof of concepts of such "FAST scans" like being able to extract heart rate and blood flow from a simple video. Techniques like EVM can even be used to detect sound from video files simply based off vibration of the objects captured in video. It seems it wouldn't be too outlandish to assume that EVM techniques + some other more classified technologies + unlimited funding of the NSA/Homeland Security/CIA/TSA could be combined to create something like the tech purported in the link. For the record, I don't work in the actual lab that developed EVM, just in a similar field.

EVM info: http://people.csail.mit.edu/mrub/vidmag/

http://people.csail.mit.edu/mrub/VisualMic/

[jimktrains2]

OK, so lets assume the cameras work. For the same reasons polygraphs have no basis in science, nor would these. I fucking hate the TSA and am fucking furious every time I have to wait to be sexually molested. I'm also worried that they'll fuck up and I'll miss my flight. I'm nervous because I'm worried. I'm constantly looking around to find out how many TSA agents are just fucking around doing nothing.

Yet, I'm not a terrorist. And you know what? most people aren't terrorists. The false positive rate so high it essentially makes the test meaningless and unable to discern an actual threat.

[hurin]

The rather strong verbal comments aside - I think this is almost certainly correct.

Polygraph tests are notoriously inaccurate - where as for something like terrorist threats the number of real positives you are interested is probably in the x > 1/10^6.

That's totally fantastic - consider that pregnancy tests have a false positive rate in the single digits.

[gknoy]

The language is strong, but pretty much encapsulates the impotent anger I feel even __thinking__ about interacting with the TSA. On the rare occasions when I must, my anxiety and fears match very closely to the GP's post's language.

[tajen]

And this is why... I don't go to US. There has been two highly interesting conferences in NY and one in SF (for which another founder offered to give me his tickets for free). But I use a false name on Facebook, which is against my contract with an american company, which is an impersonation, and I don't want to risk it with a TSA agent.

[hueving]

Your reasoning is nonsensical. The TSA has nothing to do with enforcing company contracts.

[zaroth]

The newer scanners are much less invasive. You can turn your head and see the display, it's very basic. And I've found TSA agents are much more interested in getting you through security, than holding you up in security. I saw many things that were extremely upsetting a few years ago, but barely much of anything even noteworthy lately.

Now this FAST thing anyhow, sounds insane. I will assume it just doesn't work and isn't going to be a Problem. But really, you would want to be able to stop it or put better bounds on it.

But from what I can tell, you can only sue to stop something after it has already harmed you, not just because it could cause future harm. You have to show damages to even be heard. So, how can you show damages until after it's already built and deployed?

[jimktrains2]

> The newer scanners are much less invasive. You can turn your head and see the display, it's very basic.

But we'll nicely violate your 4th amendment rights, it's OK!

> And I've found TSA agents are much more interested in getting you through security, than holding you up in security.

Which is why the assign 1 person to the first class line. 1 person to the pre check line. and 1 person to the 4 normal lines. and then have 4 people standing around doing absolutely nothing. The TSA (as an org, not as individuals) is concerned with making money for large political contributors and keeping people afraid.

> But really, you would want to be able to stop it or put better bounds on it.

But really, I want to stop everything that's happened because it's all been insane, completely unhelpful, and worth exactly $0 of the billions we've spent on it.

> So, how can you show damages until after it's already built and deployed?

By having enough people who actually care and are politically active enough to be very loud when things like this happen. Senators and Representatives need to know that their constitutes won't stand for it.

[Zigurd]

Polygraphs are not just ineffective, the practitioners are frauds and quacks. It is widespread public knowledge that polygraphs simply don't work any better than reading tea leaves or entrails. They are perpetrating this fraud on the public and on the law, and ought to be harshly punished for doing so.

[TrevorJ]

Polygraph tests do work, but as a social engineering technique and nothing more. Test takers are manipulated into behaving in certain ways based on the belief that the machine functions as they are told. The key is in the way the administrator interacts with the test taker.

In short, the same techniques don't apply to this TSA tech and so I doubt it works.

[rasz_pl]

Polygraph most certainly work ... if you believe they work while being tested.

[sitkack]

Maybe you're fed up, maybe you want to be by yourself...who knows. So you look down and see a tortoise. It's crawling toward you...

You reach down and flip the tortoise over on its back, Leon.

The tortoise lays on its back, its belly baking in the hot sun, beating its legs trying to turn itself over. But it can't. Not with out your help. But you're not helping.

I mean you're not helping! Why is that, Leon?

[Balgair]

Other posters point out that polygraphs have large error rates and that even the thought that they can 'read you mind' acts as a deterrent. In general, reading one's own mind is somewhat difficult given a stressful situation, let alone reading another's.

Yes, you can do heart rate stuff, but many medications affect that tremendously, you can do temperature stuff via IR video, but that is also super error prone. Just running to grab your flight, a perfectly reasonable thing to do at an airport, would trip every alarm in the building. Your gait would be off, your temperature, heart rate, blood pressure, eye dilation.

The issue is that humans come in an incredible array and variety of conditions. Even if you could take a baseline for every person that would ever want to fly before they got to the airport in a sterile room, the data would be useless by flight #10. This is leaving aside the tremendous human rights issues, engineering, and cost factors.

That said, computers are getting better and better at this every day, so the feasibility is there, it is just that you have to do tracking on an unheard of scale and on a day to day running basis across every single person on the planet. The disease and medical applications are the way to get it 'passed' (whatever governing body that may be), not the TSA.

[batbomb]

...and it would be readily defeated with some Xanax, and a bunch of false positives from people afraid of flying, authority figures, or missing their flight. Or any of 1000 other human reasons why people get stressed.