Slashdot Mirror
Biometric Terrorist Detector
neutralino writes "The Wall Street Journal has this story about a biometric airport security system which uses biometric responses — blood pressure, pulse and sweat levels — to series of questions ("Are you smuggling drugs?") to identify passengers with "hostile intent." According to the article, "In the latest Israeli trial, the system caught 85% of the role-acting terrorists, meaning that 15% got through, and incorrectly identified 8% of innocent travelers as potential threats, according to corporate marketing materials.""
The busiest airports in the world handle 30-80 million passengers per year. With an 8% false positive rate, a 30M/year airport would flag almost 8,800 innocent people per day, per airport as a terrorist. How can this be considered even remotely feasible? Even if getting flagged just means that you have to undergo a more rigorous personal inspection it's going to piss off a lot of passengers. Plus the TSA people aren't going to put much creedance into something that dramatically increases their daily workload, but might catch one terrorist every decade. Just another misuse of expensive technology.
Entrepreneur : (noun), French for "unemployed"
I think this is what we've been waiting for. Some method to intuitively deduce whether a person is telling a fib! The only thing this device is really missing is a name. How about ... the Polygraph? Wow, kind of catchy!
Sorry, but my blood pressure would rise if some cop comes up to me and starts interrogating me in the middle of an airport. most people almost have a heart attack when they are driving and you see the blue and red lights roll up behind you. I don't see how this is the slightest bit effective.
Remind anyone else of: "You're in the desert. You see a turtle on its back and it can't flip over. Unless it gets on its feet it will die. But you won't help it. You're going to let it die. Why is that?" (paraphrased.)
... their statistics are based on actors- who can't reasonably be expected to have genuine responses to those types of questions.
I bet there are quite a high percentage of people who, just by hooking them up to the polygraph apparatus (which is basically what we're talking about) would have elevated levels and potentially have a panic attack in some percentage of the population.
I'm betting they wouldn't even require a licensed (or certified, or whatever) polygrapher to run it, further decreasing the accuracy on an already questionable technology.
If I were a terrorist, I would pretend to be a normal person, this thing won't fly.
It reminds me of films like Airplane where the scanners stop and beat up the little old grannies but welcome the missile/gun toting libyans through.
liqbase
So when this becomes common practice, will you really be suprised when you're asked a string of questions like this?
Life is rarely fair. Cherish the moments when there is a right answer.
Bush will buy these systems that let one in six lying terrorists through, while sending one in twelve random innocents to Guantanamo. Instead of spending a mere $6M (2/1000ths of 1% of the Iraq War bill to date) on explosives detection systems.
--
make install -not war
Now my fear of flying is going to get me a cavity search. Life is just coming up roses for me...
...the idea is utterly worthless, since if you're a polished and practiced enough liar, your bodily functions are not going to change significantly, because you believe every word you're speaking. And plenty of people are going to be nervous at the types of questions, the thought that they might be lying when asked if they've used drugs or something similar when they remember the pot they smoked in college, and generally be ramped up anyway from waiting around to pass through security. It's the same process that causes your blood pressure to be higher in the doctor's office than it is when you take it at home.
GetOuttaMySpace - The Anti-Social Network
If you've ever seen a 6-foot tall crew cut tough as nails El Al employee ask you about your luggage, you know what I mean. They'll paw thru yuour luggage, pull out an orange, shove it one half inch from your nose and ask: "AND *WHAT* is *THIS*!??"
So people who weren't actually terrorists managed to generate an 85% positive rate? That would suggest that this can be easily triggered by overall nervousness (or in this case, people inducing nervousness in themselves as part of the role-playing). What is the difference between the mindset of "I need to be nervous so that I will act like a terrorist in accordance with my role" and "Oh my god, why does this TSA official think I'm a terrorist"? It's not real clear to me.
A real lie-detector test (like the polygraph) ought to be able to tell the difference between nervousness and an actual sense of having told a lie. Otherwise this is worthless.
Doesn't catching "role-acting" terrorists also imply that these people were bad actors?
This guy's the limit!
If you see anyone acting suspiciously, security can walk up to them and ask
"terroristsayswhat?"
most of them will reply
"what?"
proving that they are a terrorist.
Bingo! A solution that's just as reliable as a lie detector test...
Summation 2
then they accuse me of having high blood pressure?
there's no way out of this one, is there?
[place
If you mod me down, I shall become more powerful than you could possibly imagine.
AARP is going to have something new to talk about soon if this is the way things are going.
Considering Sen. Ted Kennedy supposedly made it on a 'no fly list' , all I can quip is 'just think of the possibilities'.
"Did you poison the quadrotriticale?"
Where were you when the voynix came?
I can just imagine him in today's society:
I have nothing to declare except my genius
Security! We have a terrorist mastermind in our midst! Get him!
Summation 2
...this tool in the right hands is effective. Israeli airlines and airports have the reputation for being the safest in the world. A big part of the reason for that is that they focus on passengers' behavior rather than what they put into their bags. Granted, the volume of air travel to and from Israel is probably a tiny fraction of what most major airports see. The questions are: (a) whether the Israelis' success is scalable to other airports, and (b) whether this device is a valuable supplement to a well-trained security team--one that can understand the machine's limitations and leverage its strengths in assessing the stream of passengers.
This looks a whole lot like a polygraph test, which has been considered in court an unnecessary breach of privacy. You can't use them for evidence and you can't use them for interviews (unless you're the FBI). So what gives us the legal precedent to use them on travelers?
Until the Diebold version comes out.
putting the 'B' in LGBTQ+
From the Article: Within a year, he predicts, such technology will be able to tell whether someone's "blood pressure or heart rate is significantly higher than the last 10 people" who entered an airport. What use is blood pressure for detecting terrorists? 16 % of people in the United States have undiagnosed hypertension. I suppose it might make for good public health screening, but I'm thinking that's a pretty bad way to detect terrorists, except perhaps those who like to binge on fast food and don't exercise...
That number is just small enough to seem effective to the bulk of opinionated political junkies who know next to nothing about computers, statistics, etc., but large enough to allow the TSA to catch no terrorists while claiming credit for being busy. It's a bureaucratic win-win. Little hard and scary work, lots of busy work and everyone is happy until it doesn't do its job when it counts, a terrorist gets through and people die.
Then, TSA gets more power.
The only time that failure is bad for such an agency is when it makes Congress seriously ask "so who is going to get the first pink slips?"
"Are you smuggling drugs?" If I was working the ticket counter, I'd ask couples "Do you cheat on your spouse?" That would provoke a much more interesting response.
I don't know about anyone else, but wheneve I fly my blood preasure goes WAY up, I sweat a lot more and I am usually VERY pissed because of the inefficiency EVERYWHERE. I'm sure I'd get flagged every time. This is total BS. I know lots of people who do the same thing, whether it's the stress fo flying, running late, flight delays, etc... I dont' see this working.
systems that identify liquids in carry-ons, systems that detect material on clothing that are common in bomb making, etc... are MUCH better options.
Putting people in a two hour long queue to go thru this system and then flagging them for being upset, sweating, etc... is just plain idiotic.
Prof. Farnsworth - "Oh a lesson in not changing history from Mr I'm-My-Own-Grandpa!"
85% of people pretending to be terrorists were identified as threats?
Sounds like an 85% false positive rate to me...
The problem with profiling is not just that it's wrong or not "PC," but that it doesn't work. Remember, the terrorists aren't dummies. If the authorities start pulling every Arab off of every plane, the terrorists groups will soon get wise to that. They will search their ranks for the least Arab looking members or recruit radicalized westerners. Narrowing the focus of your search just gives your target a chance to adapt.
I may be in the minority, but I KNOW my BP/heart rate would rise just from fear of the machine not getting it right:
'What if it flags me, would they do the cavity search? would I miss my connection? would I have to spend a night in a packed jail with real criminals? would I be held until my next bowel movement?' (which they do with pregnant women 'drug mules' that cannot have Xrays)
And I rather suspect the terrorists that have daily polygraph training sessions would pass with no problems.
Obama likes poor people so much, he wants to make more of them.
It is truly bizarre that someone who is smuggling drugs is grouped in with terrorists as having a "hostile intent". True, both are breaking laws, but drug smugglers have a motive for being non-hostile - they don't want to get caught. They just want to get from point A to point B without interference. Which, paradoxically, gives them the same motive as the TSA.
Hmmmm...that gives me an idea. Drug smugglers could be useful allies in the war on terror. I suggest a new TSA policy. Let one dealer through on each flight. Grant him the right to carry, say, 5 kilos of drugs exempt from the law. Let him also carry a gun - uh, no - REQUIRE that he carry a gun as part of the deal. You can be sure he will not let a plane get highjacked without a fight. And a terrorist organization would think twice about highjacking a plane - even if they could overpower the dealer - knowing that the Medellin or some other international drug cartel would then be out for revenge.
Not only would flights be safer, but this is a very profitable policy for the TSA; they save the cost of hiring air marshalls, and the dealer would pay a bunch of money for the privilege.
What the heck - let's take this idea to its logical conclusion. Let the cartels run their own flights. I'd feel safer on Medellin airlines that I do on American or United, etc. ( I'll bet that they could also put the fun back in flying: "Would you like some coffee, senor? Cocaine? Hashish?" )
Not sure whether that would work - Al-Qaeda seems to have enough resources to circumvent racial profiling. There has already been the case of that shoe bomber, and there was Anne Murphy (that pregnant Irish girl), too.
Assuming the marketing literature is right about the detection rates, that alone renders the device useless. Consider this: for every terrorist there are at least a thousand legitimate travellers in any given airport. At an 8% false-positive rate, you'll incorrectly tag 80 innocent travellers during screening. Assume you tag the terrorist as well. You've now got a group of 81 people, 80 of whom are innocent. What's the public reaction going to be when, after the delays and the hassles to all those people, it turns out that 98.77% of the time your "detector" is wrong? And this is conservative, assuming a low number of travellers and a high percentage of terrorists. It wouldn't suprise me if a major airport like Heathrow handled several tens of thousands of travellers every day and only saw any terrorists at all on one day a month if that often.
"In the latest Israeli trial, the system caught 85% of the role-acting terrorists, meaning that 15% got through, and incorrectly identified 8% of innocent"
Assume say 2,020 people. 20 are terrorists.
The machine will identify .08 * 2000 plus 20*.85 = 177 people called terrorist by the machine
Of those only 17 are really terrorists (less than 10%), the rest are innocent. 90% wrong decisions
Of the people called 1843 "innocent" by the machine, 6 would real be terrorists. Less than 1% wrong decision there, but even 6 are 6 too many.
This machine looks to do nothing but provide a false sense of security, while causing MAJOR trouble for a huge number of innocent people.
This is basically just a Lie detector, used for a VERY bad methodology. Lie detectors ARE usefull, if used correctly. Specifically you use them to confirm knowledge, not motive.
I.E. "Lie detectors" can NOT detect lies said by the suspect, they detect Nervousness. The proper way to use them is simple. Say you have a woman killed when someone cut her throat. You take suspect, before he has seen the body, or heard anything about her murder and you ask him:
1 "Did you blow up the victim?"
2 "Did you cut the victim's throat?"
3 "Did you shoot the victim?"
4 "Did you run the victim over in a car?"
If the man is innocent, he will be no more nervous on question #2 than the other questions. If he is guilty, chances are question #2 will cause a HUGE jump in nervousness, as compared to the other questions.
Even this is not fool proof (if the suspect happens to be afraid of knives/was cut by a mugger, bad results are likely), but it is certainly a lot more helpfull than the standard practice.
excitingthingstodo.blogspot.com
Passengers swip their passport into the machine. First off, I seem to remember that all of the 9/11 planes were on domestic flights and therefore people wouldn't have their passports.
Secondly there was just recent concern about ease of duplication of RFID passport data. I hope no one decides to put this technology in use until alot of problems are worked out.
Just how accurate is "role acting" terrorists? An 8% false positive rate is almost 1 of every 12 people. Perhaps a role of a twelve sided die would work as well.
Think Deeply.
"I would much rather be safe than happy on a flight"
For sure but too much of a delay and it's just not worth travelling. London to Paris is 1 hr 15 minutes (approx), right now we're being told minimum 2 hours wait time to get on the plane for European short hop flights. It's one thing to queue for 3 hours for your London -Australia holiday flight but another if you want to get somewhere across Europe, have a meeting and fly back the the same day. Luckily I don't have to do this any more but a lot of people do.
I'm off to Copenhagen next week for a conference from London, 2 hour flight. A two hour wait in the airport I can just about cope with but when they start talking about 3,4,5 hour delays, heck, you got to wonder if it's worth it. Maybe we just have to rethink about how we do business in remote locations. I reckon the train and ferry companies are probably pretty happy right now, and the videoconference people as well!
No system like this will ever work for a very simple reason - one well known to those of us who do medical screening.
Note the following.
Assume 1 in every thousand passengers is a terrorist - surely a wild exaggeration. So the probability that a random passenger is 'innocent' is 0.9999.
This sytem will 'catch' 85 terrorists and 7,992 innocent people from every 100,000 passengers.
So what we call the 'Negative Predictive Value', that is the probability that someone who passes the test isn't a terrorist is not too bad at 0.99984, down from 0.9999 before we did the test,
The 'Positive Predictive Value' is a truly uninspiring 0.01052. Just over 1% of those who test positive truly are actors pretending to have evil intent.
For a more realistic 1 in ten thousand passengers being a terrorist we get 8 terrorists and a total of 7999 'innocent's. This assumes that actors mimic terrorists well.
This would stop international flights dead, but buy shares in it anyway.
-- Anthony Staines
My take is that will filter out most of your terrorists though.
Really? You think that a system that produces thousands of potential terorrist suspects a day at an airport like O'Hare is going to be treated seriously enough that it actually catches the one real terrorist that crosses through it every few months?
Um, no. The way we filter out most terrorists is by searching for physical evidence that actually indicates intent and which makes false positives easy to identify. That we have already accomplished. The only way this buzz-word-wrapped version of a shitty polygraph is going to make planes a less attractive target for terrorism is by ensuring that nobody wants to fly so there isn't a target in the first place.
The enemies of Democracy are