Slashdot Mirror
Researchers Design Bot To Conduct National Security Clearance Interviews
meghan elizabeth (3689911) writes Advancing a career in the U.S. government might soon require an interview with a computer-generated head who wants to know about that time you took ketamine. A recent study by psychologists at the National Center for Credibility Assessment, published in the journal Computers and Human Behavior, asserts that not only would a computer-generated interviewer be less "time consuming, labor intensive, and costly to the Federal Government," people are actually more likely to admit things to the bot.
Eliza finds a new job.
That's good, now tell me about your mother Dave...
From the Vice article, this sounds a lot like a robotic polygraph - the article directly mentions using "electrodes to measure cardiographic and electrodermal responses".. which is essentially what a polygraph does. I can't imagine that a robot will be any more effective at applying baseless pseudoscience than a human would - in fact, I wouldn't be surprised if the number of people who have their careers ruined due to a failed polygraph is higher with the robot than with a human "interpreter".
Wouldn't it be much more efficient to just eliminate the polygraph altogether?
We already know they're robots. You, we're not so sure of.
Sounds more like M-M-M-Max Head-Head-Headroom, to me.
Holden: You're in a desert, walking along in the sand, when all of a sudden you look down...
Leon: What one?
Holden: What?
Leon: What desert?
Holden: It doesn't make any difference what desert, it's completely hypothetical.
Leon: But, how come I'd be there?
Holden: Maybe you're fed up. Maybe you want to be by yourself. Who knows? You look down and see a tortoise, Leon. It's crawling toward you...
Leon: Tortoise? What's that?
Holden: [irritated by Leon's interruptions] You know what a turtle is?
Leon: Of course!
Holden: Same thing.
Leon: I've never seen a turtle... But I understand what you mean.
Holden: You reach down and you flip the tortoise over on its back, Leon.
Leon: Do you make up these questions, Mr. Holden? Or do they write 'em down for you?
Holden: 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 without your help. But you're not helping.
Leon: [angry at the suggestion] What do you mean, I'm not helping?
Holden: I mean: you're not helping! Why is that, Leon?
[Leon has become visibly shaken]
Holden: They're just questions, Leon. In answer to your query, they're written down for me. It's a test, designed to provoke an emotional response... Shall we continue?
Can you make smoke come of it by saying that everything you say is a lie?
So when an idiot gets clearance, there is no one to blame? Remember the finger pointing after Snoden (not that he's an idiot) No more of that blame game with this! Problem solved.
"Are you now or have you ever been a member of the Communist Party of the United States..."
Whoopsie, wrong questionnaire.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
This is a murky field. A polygraph does present useful information; it's just not necessarily whether the person is telling the truth. The major decision part of any polygraph system is the operator, and they need to have sp,e training in physical psychology to be predictably any good at using the equipment.
Seems to me that this new system falls into the same category. They'll be able to get some new data that would have been obscured before, but the interpretation of the data will still require an expert.
Personally, I think this is better than leaving it up to a human, as the human mind has known defects during the data acquisition phase -- these systems don't have those weaknesses, and while they can't draw any conclusions, they gather a different (and in some cases more complete) set of information than a human by themselves would gather.
The problem comes when people conflate the results of the tests with factual certainty -- both systems require interpretation, and as we all know, statistics lie 99.8% of the time.
Surely you can point us to a double blind study to quantify lie detector effectiveness? They don't exist?
Yes they do. The Wikipedia page lists several. What they find is that polygraphs work better than chance, but below perfection. They certainly don't provide the level of "beyond a reasonable doubt" required in a criminal court, and they can be fooled by a someone trained to deceive them. But for most people, they work most of the time. That is good enough for their use as a first level screening device. You would be an idiot to blindly accept their results, but you would also be an idiot to ignore the results completely.
It is an effective prop. But only for the uninformed.
Wrong. It takes more than just being "informed". To trick a polygraph takes training. So how many moles hoping to infiltrate the FBI are going to respond to a Craigslist ad for "Polygraph Deception Training"? Guess who places those ads.
Let me tell you about my mother. [blam!]