TSA Behavioral Detection Statistics

Interesting data from the U.S. Government Accountability Office:

But congressional auditors have questions about other efficiencies as well, like having 3,000 "behavior detection" officers assigned to question passengers. The officers sidetracked 50,000 passengers in 2010, resulting in the arrests of 300 passengers, the GAO found. None turned out to be terrorists.

Yet in the same year, behavior detection teams apparently let at least 16 individuals allegedly involved in six subsequent terror plots slip through eight different airports. GAO said the individuals moved through protected airports on at least 23 different occasions.

I don't believe the second paragraph. We haven't had six terror plots between 2010 and today. And even if we did, how would the auditors know? But I'm sure the first paragraph is correct: the behavioral detection program is 0% effective at preventing terrorism.

The rest of the article is pretty depressing. The TSA refuses to back down on any of its security theater measures. At the same time, its budget is being cut and more people are flying. The result: longer waiting times at security.

Posted on April 20, 2012 at 6:19 AM • 32 Comments

Comments

VincentApril 20, 2012 6:40 AM

Israel El AL airline has been using these techniques. It seems to have been effective, because El Al has been a target for long, and only one El Al airplane have been successfully hijacked (according to Wikipedia). (See also on CNN: How the Israelis do airport security)

Joe WhiteApril 20, 2012 7:02 AM

Wait a minute, Bruce. You've said in the past that you're a big fan of behavior detection, but here you say it's 0% effective? I must be missing some context.

CuryworksApril 20, 2012 7:21 AM

If you have been through Israeli security you will a personnel quality difference. Also does the avg. TSA worker even believe in a terrorist threat.

SpeedApril 20, 2012 7:29 AM

That works out to 17 passengers "sidetracked" per officer per year. Less than 1.5 per month -- all false positives.

Another KevinApril 20, 2012 7:43 AM

I believe the second paragraph. It doesn't say that we've had six terror plots. It says that we've had six allegations of terror plots. But don't believe the allegations.

(Besides, with the broad definition of "terrorism" in the laws today, I suspect that we've had considerably more than six terrorism convictions - to say nothing of plots. Can't any crime be labeled a terrorist act if the prosecutor is sufficiently clever?

KennyApril 20, 2012 8:07 AM

The TSA can't back down on its security theater – someone else has to give them cover to do so, and it's not much of a political opportunity. No one person feels that much of the cost (pain), tho I personally seethe every time I fly now, but imagine the reaction *were* a single terrorist act to occur that is in some way related to a commercial flight ...

BobApril 20, 2012 8:17 AM

I'm glad airport lines are getting longer. I will arrive early, and then I will do my very best to back up the security line even more when I get to the airport as I always do.

The best way to end the TSA is to make going through airport security so slow and inconvenient that people actually start standing up and objecting to these ridiculous and obscene displays of security theater.

Freedom is more important to me that convenience. It should be to you too.

kingsnakeApril 20, 2012 8:50 AM

Does it detect behavioral problems within the TSA ranks, like stealing $300 from an Air Force veteran?

VatosApril 20, 2012 9:57 AM

You say "the behavioral detection program is 0% effective at preventing terrorism" and discount what they are doing.

But you recently praised Linode.

Why not judge Linode by the effectiveness of their ideas too? If there is no effectiveness then Linode Password Security efforts should be discounted as theater as well

dragonfrogApril 20, 2012 10:00 AM

I'm sure that if we correctly labelled as "terrorism" all the things that really qualify (i.e. not just brown people blowing up white people, but also white people firebombing religious buildings attended by brown people, etc.) then we had well over six terror plots in 2010.

Arnaud PalissonApril 20, 2012 10:06 AM

I think you can't say that behavior detection (BD) is 0% effective.

First, there was no terrorist attack from any US airport since 9/11. So, objectively speaking, behavior detection is at least as effective as other methods.

Moreover, El Al did detect Richard Reid when boarding a flight in July, 2001 in Amsterdam (he was there to gather information for an eventual attack).

The same Richard Reid was detected twice (on Dec. 22 and the day before) in Paris-CDG by French profilers, when trying to board the AA63 flight he wanted to blow up. (he finally boarded the plane the second time, due to a final decision from French Police or the airport authority).

In 1986, in London-Heathrow, Anne-Marie Murphy got detected by an El Al profiler (and she even ignored she had a bomb in her suitcase!)

In 1999, Ahmed Ressam got detected by a CBP officer at Canada-US land border thanks to behavior detection.

And so on.

The problem with BD at the TSA lies in the way TSA adapted the technique. Instead of using the Israeli method, they mixed it with several psychology theories. For instance the infamous "micro-expressions".
The result was an incoherent and inadequate system : in its first version (2003-2011), the TSA used BD without even asking questions to passengers. Inadequate but NOT ineffective.

And in its second version, TSA BDOs now ask questions to passengers. It's better. But still inadequate because the questions are always the same. That means that it's easy for the bad guys to trick the system with a simple cover story.

The Israeli system goes much further in security questioning, And that's where TSA has to be heading, for "Behavior Detection 3.0".

And when the second paragraph explains that BD didn't not detect the 16 individuals later involved in 23 attacks, at the time they boarded a plane in the US... Gee!... they were not going to blow up that plane : at that time, they were just passengers, flying from Point A to Point B. Thus they couldn't be detected for having the behavior of a terrorist trying to blow up the plane : they were NOT going to blow up the plane.

Frank Ch. EiglerApril 20, 2012 10:21 AM

"We haven't had six terror plots between 2010 and today. And even if we did, how would the auditors know?"

Perhaps their access to TSA/FBI information is simply better than Bruce's.

Robert ThilleApril 20, 2012 10:55 AM

The statement doesn't specify that the terrorist plots were local to the US, only that those involved traveled thru our airports.

H. HolzerApril 20, 2012 11:02 AM

Do I see trolls here for the TSA? Seems apologies for them are oozing from somewhere, at an usual rate. I think 0% is quite fair. Why? Because you have to account for the sum of terror the TSA themselves have inflicted on innocent civilians, bakers, children, elders, and to the image of travel in this country. Now they'd have to stifle a few more than six plots to compensate for that!

AbdulApril 20, 2012 11:37 AM

Actually, you can thank my magic stone for the lack of terror plots. I acquired it at the same time as the TSA deployed their BD agents. The lack of any plots since that time is clear evidence that my magic stone works.

I will be acquiring other magic stones soon, and will secretly place one in each airport I visit, to further increase the prevention of plots. No need to thank me, I'm just doing my duty as a loyal citizen and believer in magic stones and TSA.

anonymousApril 20, 2012 12:26 PM

An off topic link for you, you may already know

GHCQ has released some of Turing's Bletchey park papers...

http://www.bbc.co.uk/news/technology-17771962

Unfortunately, they're not online. National archives in London. Perhaps someone with your....audience could arrange to get them /actually/ published?

GeorgeApril 20, 2012 1:33 PM

This sounds like it's referring to the GAO report released in 2010. The report essentially declared the BDO program as implemented by the TSA a complete waste of over a billion dollars. The TSA apparently put it in place without testing, with no valid evidence for the efficacy of either behavior detection or the way the TSA implements it, and without even a plan to verify it.

The part about the 16 individuals allegedly associated with terrorist plots was a retrospective review of records. They found that those people made 23 trips through eight airports where BDOs were deployed, but there were no reports of any BDO picking up on their "signals" and interrogating them.

This review was secondary to a more comprehensive review of other individuals who were "spotted" and referred to law enforcement. Between 2004 and 2008, BDOs questioned around 152,000 passengers. Of those, 14,000 were suspicious enough to refer to law enforcement. Of those, 1,083 were arrested, with charges relating to drugs, immigration, outstanding warrants, and fraudulent documents. None of those were found to have any connection with terrorism or threat to aviation. Thus, for the billion-plus dollars SPOT is costing, we get a system that provides a 0.7% detection rate for (possible) criminal activities that do not threaten aviation security. John Pistole might insist it's worth it, but I doubt anyone else would consider it good value for money.

Besides providing the only published objective cost-benefit analysis of a TSA program, what I found most interesting about the GAO report was that it included the DHS response to the audit. The GAO made a number of recommendations, the most notable one calling for an independent assessment of behavior detection and SPOT by experts outside the DHS.

Most of the DHS response was a defense of the TSA and DHS and a denial of the GAO claims. They insisted that the GAO results were invalid because the auditors were unqualified, as the only people who are qualified to assess the TSA already work for the DHS. So they absolutely refused any independent assessment, but would have their own "Science and Technology Directorate" review the results and determine what to do. And while they "concurred" with the other GAO recommendations, the wording of the "concurrence" implies that they intend to ignore all of it. And, of course, Blogger Bob immediately responded to the report's publication with a blog post effusively lauding the SPOT program and the BDOs, and completely ignoring the report itself.

Since nobody is following up on this (or any other) audit, or holding the TSA accountable for anything, DHS can, as always, get away with telling the GAO to go suck an egg. That, of course, is the root cause of the TSA's many problems.

GeorgeApril 20, 2012 1:53 PM

@Arnaud Palisson: The problem with BD at the TSA lies in the way TSA adapted the technique. Instead of using the Israeli method, they mixed it with several psychology theories. For instance the infamous "micro-expressions".

The TSA can't use the "Israeli method." The Israelis have a complete and coherent system in which all airport employees are involved. Their equivalent to BDOs are highly trained and highly competent, and are allowed to use their intelligence and instincts to observe and ask questions. But other airport employees are also on the lookout for "hinky" behavior, and can call in the experts as necessary. It also helps that everyone has a personal stake in security.

The TSA obviously can't do anything like this, given the quality of the employees they have available. Like everything else, they can only create a sham appearance of security. Blogger Bob is very proud of the full week of special training BDOs receive. That may be enough to teach "officers" how to elicit stress responses in passengers who have a well-justified fear of the TSA, thus producing "numbers" that supposedly verify the effectiveness of the system. But there's no way it can be effective, even if you define "effectiveness" in terms of numbers and false positives.

And even if accept on blind faith the TSA's claims about BDOs being experts at detecting "stress" signals, the environment the TSA and airlines have created would make it useless. If the majority of passengers are stressed due to the nature of flying and the well-justified fear of the TSA, how could even "expert" officers pick out a "signal" from a terrorist from such a high level of "noise"?

And in its second version, TSA BDOs now ask questions to passengers. It's better. But still inadequate because the questions are always the same. That means that it's easy for the bad guys to trick the system with a simple cover story.

And what happens when a passenger is hesitant or fearful of engaging in a "friendly chat" with a TSA officer, not because he's a terrorist but because he's justifiably afraid of being hassled, detained, or humiliated? Here again, the TSA's own "poisonous" approach to passenger screening undermines the effectiveness of even valid security measures.

I'm no expert on security. But it seems obvious to me that the TSA would be much more effective if they earned the trust, respect, and cooperation of the public. But it's just as obvious that such an approach is completely alien to the groupthink mindset of "law enforcement" and "security," that apparently holds as an article of faith the belief passengers need to be treated like enemies and bullied into submission.

Brandioch ConnerApril 20, 2012 2:45 PM

What Bruce said was:
"But I'm sure the first paragraph is correct: the behavioral detection program is 0% effective at preventing terrorism."

He did not say that behavioral detection was 0% effective.

Just that the program of behavioral detection that the TSA has implemented is 0% effective.

The TSA has not caught a single terrorist yet.

Mark BennettApril 20, 2012 4:40 PM

After a little more online digging, it appears to me that these are numbers from a May 2010 GSA report. There are newer reports, but I haven't dug in to them to see if they tell a different story.

Kevin MacnishApril 23, 2012 7:11 AM

There are several problems with drawing conclusions from the information in the GAO report as it is, and even more so with all behavioural detection on the basis of the TSA's SPOT programme (excuse UK spellings) which is being discussed here.

Firstly, and leaving aside definitions of terrorist, was SPOT attempting to recognize all terrorists or just those about to plant a bomb? The latter seems more likely (why should a terrorist per se act any differently in an airport from anyone else?). If the 16 who passed the SPOT checks were acting suspiciously, then it failed. Otherwise not. Ultimately it can only recognize people acting suspiciously, not any and all terrorists.

Secondly, was SPOT supposed to be the sole means of discovering terrorists? Usually you have to accept a margin of error and use a complex of methods to allow for false negatives. If you want a system with no false negatives then prepare for intense one-on-one interrogations every time you fly.

I'm not an apologist for SPOT but do think that a lot of conclusions are being drawn to hastily regarding its efficacy, both on this site and generally elsewhere.

I've written quite a bit more on SPOT here: http://www.springerlink.com/content/...

Arnaud PalissonApril 23, 2012 7:59 AM

@ Kevin Macnish -
Absolutely right. I’ve been telling that for months (years?).
But, well, it’s written in French, so...

@ George -
The word method I used may be inappropriate. Let’s say Israeli concept. You are definitely right about the quality of TSOs and I’m not suggesting to transfer the whole Israeli method to the US. That is just impossible. I’m aware of that. But you can’t be against an improvement of the system. You just can’t say, like Janet Napolitano during a former trip in Israel: It won’t work in the US. Period.

In several cases (like the ones I gave in my previous comment), it didn’t take a whole redesign of the security system to detect Ahmed Ressam and Richard Reid. It proves that such a system would work in US airports.

Now, about the trust and the stress, the Israeli sytem is totally able to make the distinction between people who have something to hide for illegitimate reasons and people who are nervous about talking with soldiers or police officers at airports. I don’t see why it wouldn’t be the case in US airports. Profiling is not just about detecting stressed people.

You are also right about the indicators TSA uses to prove the system works: numbers and false positives. I know the problem, believe me. I teach security profiling to the Montreal Airport Patrol constables. And I've been asked to get statistics from the field afterwards. But how do you prove with numbers that someting works when there is nothing to count? And that’s also what happens with the US press and politicians. They want TSA to catch terrorists on a mission. But what if no terrorists on a mission ever passes through US airports? Does it mean that the system is wrong? Nope.

Sure, TSOs are mainly taught to use their reptilian complex, dedicated to the strict respect of procedures.
And yes, it will take a huge change of mind.
But does it mean that it will never change?
I think there is room for improvement. Even at the TSA.
You may say I’m a dreamer, though.

TiredApril 25, 2012 7:20 AM

It's not going to stop until we stop flying. Maybe we could boycott all flights on Wednesdays or something.

Not only is it not stopping, but it's spreading. Flying out of Hong Kong on Monday, they would not let me take my nail file through security. I had, and took, the option of leaving security to check my

TiredApril 25, 2012 7:23 AM

Sorry, message truncated.

nailfile, but going through security a second time, they had serious concerns about my crochet hook... Flying out of Tawian last summer, they confiscated my blunt-ended scissors. The only improvement is that in Hong Kong, Taiwan, and Japan (at least), removing shoes is no longer required or requested.

Jack MorganApril 26, 2012 11:12 AM

The story about the pat-down of a 4-year-old girl at Kan. Airport is really interesting. What caught my attention was the way in which the TSA defended its actions. In their response, they indicate agents followed “proper procedure”. It seems to be their magic answer to why these agents seem to be acting so thoughtlessly. Why of course they are… they are following “proper procedures”. Ta da!
Really?
If TSA agents are supposed to behave like industrial robots, then why is the government not replacing them with real, metal and silicon based industrial robots? It would save millions and we would achieve the same results as human screeners. Robotic systems would follow procedure, there would be no thought and no questioning and we would not feel that the security screening procedures are ridiculous because there wouldn’t be an expectation of human intelligence involvement on the TSA side. Just automation – which is exactly what is happening today, except its flesh and blood agents with salary and health benefits.
The thing that really worries me is that having human automatons following procedures so blindly gives the real adversary, the extremist terrorists - an unquestionable advantage. They know the process is going to be followed without thought. They know there is no independent thought even allowed and no ability on the part of the TSA agents to respond creatively to exceptions and unscripted situations – that is exactly the weakness that they will exploit. Defeating a robot is not hard to do… flesh and blood or metal makes no difference.

Lord NikonMay 3, 2012 7:38 PM

You people are so blind,that you don't ever see any other threats that are out there. On a daily basis there are other things going on around you and the only thing you have is tunnel vision. Before 9/11 happened there was absolutely no protection from somebody doing harm to you, now at least there is something implemented. Its also a shame to think you narrow minded people want to see thousands loose there jobs, not to mention, that a great percentage of them are military veterans. I am happy that TSA is in our airports and am astonished at the good job that they to do protect whiners such as yourselves. There will always be lines in airports and even when I traveled before TSA, there we ridiculous lines and stupid security. All of you are just weekend patriots with nothing else to do but bitch on a blog, GET A LIFE, move on. TSA is here to stay.

ice purveyorAugust 18, 2012 10:08 PM

I'm thinkin troll. TSA screeners aren't LEO and taking away over 3oz of dangerous hair gel wouldn't qualify them to be in the know of those "things going on". It's unfortunate that you don't mind some strangers hands on your wife or little kids. Air travel now is something you have to endure.

Israel uses highly trained officers, not part time soccer moms.

Dave JSeptember 8, 2012 8:41 AM

I fear the TSA more than flying. I don't think it is the business of the TSA to ask me where I plan to travel, what I plan to do on my trip, where I plan to stay, when I plan to come back, or anything else regarding my life. Terrorist attacks were rare occurrences on US soil even before 9/11 and the argument that we haven't had a single attack on US Soil since 9/11 cannot be attributed to the TSA or any post 9/11 change since there really aren't many terrorist attacks on US Soil in the first place. Prior to 9/11 the last terrorist attack on US soil was the Oklahoma City Bombing which happened in the mid 1990s. I've read about racial profiling from the TSA SPOT program but what about people who may have abnormal facial tics or expressions such as people who suffered a stroke or have a neurological or psychiatric problem? People with abnormal facial expressions or tics would probably be easily targeted by the TSA SPOT program as potential terrorists because they don't behave "normally". The TSA claims they can tell the difference between someone who is afraid to fly and is nervous due to this but what about people who have anxiety disorders? These people would also be flagged as suspicious especially when they realize a TSA official can have them detained and deny them the right to fly for any reason making even people without anxiety disorders nervous. You're more likely to die in a plane that has crashed due to mechanical problems than one that was hijacked or had an explosive device on board! For some reason we have given the TSA the green light to further intrude in our lives and even if you don't fly you will soon see the TSA at VIPR checkpoints that will be at sporting events, random road blocks, and other places in addition to the traditional bus and train stations. All because of 1 terrorist attack that the US Government had warnings about but chose to ignore. After all, Al Qaeda already had tried to bomb the World Trade Center once in the mid-90s and is suspected of attacking targets outside of the US such as the USS Cole.

Leave a comment

Allowed HTML: <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre>

Photo of Bruce Schneier by Per Ervland.

Schneier on Security is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc..