The TSA agent can always fall back on "I'm following orders". Say what you want about that being an affirmative legal defense, it's certainly a psychological defense.
But then, you're assuming that openly authoritarian control would be reviled and cause rebellion. You're not taking into account the possibility that nobody gives a shit because their three-meals-a-day-and-Monday-Night-Football aren't threatened.
And you're assuming that the State cares what people think. Some Soviet dissident was quoted as saying "we decided that we would just denounce everyone, everyone we could think of, whether they were actually traitors or not, whether we actually had evidence or not. Surely, we thought, the State would not murder or imprison thousands of people with neither cause nor due process. We were wrong."
"I only got to third base with the girls in question and did not actually lose my virginity until I was 16. I dry-humped them, though...what does this have to do with my cats?"
"...the head of the climate unit at the University of East Anglia almost had to resign, but no wrong doing was found in the investigation."
Ho, ho, ho. The Bush administration had lawyers' statements that torture wasn't wrong.
PS last year my niece was 3. This year she is 4. Since her age is therefore increasing at 33% every year, I can confidently--and with cold, hard math backing me up---declare that in the year 2025 she will be 225 years old.
"There’s no surer way to permanently implant a true aristocracy than to elect only those wealthy enough to fund their own campaigns."
You mean like people who get to use tax money to fund their campaigns?
It's not like those poor poor Democrats had to scrimp and save and beg in the street. Brown and Boxer matched Whitman and Fiorina ad-for-ad, at least as far as I saw from watching TV.
And it doesn't help when authors who genuinely attempt to include non-white perspectives and influences are told that it's "cultural appropriation" and they're being even more racist.
In the Two Towers Extended Edition, there's a scene where the Huorns (note spelling) appear outside Helm's Deep, and the Uruk-Hai army runs into them and is wiped out.
And this matches the books; the Huorns didn't break the seige (that was Gandalf and the Eodred) but they did wipe out the fleeing Uruk-Hai army.
Yeah; it's probably one of those "if you're rational, then you'll agree that a benevolent tyranny is the most-efficient form of government that guarantees the least amount of human suffering, and you'll also agree that a superintelligent AI would be a benevolent tyrant, therefore you'll let me out of the box. If you refuse then you're irrational and therefore stupid. And you're not stupid, are you?"
Again, don't forget that the AI player can invent anything he wants. If the gatekeeper says "prove that you'll be a benevolent tyrant and that the world would be better", the AI player can say "okay, here is a complete and logically-sound proof (gives proof)" and per the experimental parameters that proof would have to be treated as valid and true.
Although this depends on the gatekeeper player being smart enough to follow the chain of reasoning. I remember someone on USEnet complaining that it was paradoxically difficult to fool people with brainy logic traps because people were too stupid to understand the logic!
*****
And, ultimately, the experiment isn't about the specifics of the transcript; the experiment is about disproving the statement "no smart human would ever allow an AI out of its box".
It's like reading breathless Fox News exposes about Stuxnet. "It's an AI Cyber-Bomb that bridges the air gap and uses Zero Day Exploits!" Um...any security professional already knows about everything that Stuxnet supposedly did, and anyone familiar with the processes involved in nuclear-material refinement knows that you can screw that up without needing any haxoring of Gibsons.
Same thing with articles about airplanes read by aerospace engineers or Popular Mechanics readers. And articles about virtually any physics experiment--"so this experiment you're doing, it can make time travel possible and also blow up the universe?"
Actually, it was JFK building on what Eisenhower had started. If it hadn't been for ballistic missiles and nuclear weapons and spy satellites, we'd have never had the technological base that made JFK think we could put men on the Moon.
One thing to keep in mind is that the LotR novels weren't colorblind. There were plenty of non-whites; they were the bad guys, "swarthy" Southrons and Easterlings.
"... if the Gatekeeper says "Unless you give me a cure for cancer, I won't let you out" the AI can say: "Okay, here's a cure for cancer" and it will be assumed, within the test, that the AI has actually provided such a cure."
And...
"The results of any simulated test of the AI shall be provided by the AI party."
So, in other words, the AI player can invent anything he wants, and it must be assumed that whatever he invents actually exists and is actually doable. So the AI player just says "I'm a super-smart AI, I've invented an energy-to-matter conversion device that will eliminate material scarcity, I have done the tests to prove it works (and the parameters of this experiment require that you accept these tests as truth), let me out and I'll give it to you." Then it's two hours of "you're a miserable bastard because you want humans to suffer and die, the only way to prove you aren't is to let me out".
"The first nine comments here all failed to get at Mark’s question, which was about why political partisans were giving slack to “their guy” on an issue where they’d berated the opponent, while attacking “their guy” on what is almost indisputably a more minor issue."
...so "is it happening to someone else or is it happening to me" isn't an answer to that question?
...so if it costs a trillion dollars to lower the defect rate from 1000 DPM to 20 DPM, that's worth it because you're eliminating 980 DPM and 980 is a really big number?
Seriously, the first question I'd ask in response to your philosophy is "what's the per-saved-unit DPM-reduction cost?" Because if the cost wipes out the profit on the saved units then it's not worth doing.
First off: "That a rational actor working inside the system’s rules might prefer — and even be able to bring about — a weak economy should scare us, even if we don’t believe they’ll purposefully try and do it."
It's funny to see the Democrats get to the place the Republicans were in 2008--"Bad People are intentionally making the economy go in the crapper so that they can use economics as a stick to beat on Republican incumbents!" I only wonder why Klein didn't address this history. (Indeed, he claims that there's "some evidence that the economy performs better under unified government". Like that's *not* what we had from 2001-2006?)
******
MFarmer: I agree. This whole thing seems like a fancy, extra-wordy version of the ad hominem fallacy.
"There will never — precisely never — be another airplane takeover using box cutters. "
But is that because "takeover using box cutters" was a one-in-a-million random occurrence? Or is it because potential attackers know we won't let box cutters get on the plane? (Even though passengers know it's necessary to fight back now, I wouldn't want to depend on a bunch of retirees and children to take down a group of suicidal knife-wielding fanatics.)
Yes? I'm not saying that it's *not* an indefensible argument?
And yes, the FDA is entirely willing to cost something off the face of the earth rather than ban it outright.
But, y'know, I can't really blame them. Despite what I said up-thread about Scary Large Numbers, if a reduction in the overall defect rate is possible then it should probably be employed. The real lesson here is the one that everyone should learn from even the most cursory reading of statistical analysis--that you can't make a decision based on just one number. "It's a really big number so it's good" is just as wrong as "it's a really small number so it's bad".
See, the thing is, I'd still look at that and say that the limit of the defect rate with the current process is 1%. The question here is not "how do we make that zero", the question is "can our business work with a 1% defect rate?"
You're doing exactly what Hanley does but from a different direction. He says "the possible improvement is small, therefore we shouldn't spend resources improving it." You're saying "no, the possible improvement is LARGE, therefore we SHOULD spend resources improving it." You're both making arguments based on the same numbers; you're just disagreeing about which way to look at the numbers.
That said: having worked with the FDA and other government agencies, they'd consider a 99.92% to 99.99% improvement to be well worth the cost.
On “Exceptionalism, Imperialism, and the Necessity of “Closed Systems””
"What answer can they offer instead? Perhaps a Euro-merican exceptionalism. A re-dressing of the the U.S.’s “Western mantle” in more PC terms."
"Take up the metrosexual post-colonial college-educated man's burden", you're suggesting?
On “The crazy misadventures of the TSA”
The TSA agent can always fall back on "I'm following orders". Say what you want about that being an affirmative legal defense, it's certainly a psychological defense.
On “Doctor Science on Conspiracy”
But then, you're assuming that openly authoritarian control would be reviled and cause rebellion. You're not taking into account the possibility that nobody gives a shit because their three-meals-a-day-and-Monday-Night-Football aren't threatened.
And you're assuming that the State cares what people think. Some Soviet dissident was quoted as saying "we decided that we would just denounce everyone, everyone we could think of, whether they were actually traitors or not, whether we actually had evidence or not. Surely, we thought, the State would not murder or imprison thousands of people with neither cause nor due process. We were wrong."
On “A Tribute to Sean Carasov”
The worst part is going to be watching all the Scientology fucks congratulating themselves over this, as though they personally pulled the trigger.
On “R.S. McCain accuses me of being a violent militant”
"I only got to third base with the girls in question and did not actually lose my virginity until I was 16. I dry-humped them, though...what does this have to do with my cats?"
It shows how you're obsessed with pussy.
(ha ha ha, ZING!)
On “Speaking of plagiarism…”
"...the head of the climate unit at the University of East Anglia almost had to resign, but no wrong doing was found in the investigation."
Ho, ho, ho. The Bush administration had lawyers' statements that torture wasn't wrong.
PS last year my niece was 3. This year she is 4. Since her age is therefore increasing at 33% every year, I can confidently--and with cold, hard math backing me up---declare that in the year 2025 she will be 225 years old.
On “Best 2010 Election Results News”
"There’s no surer way to permanently implant a true aristocracy than to elect only those wealthy enough to fund their own campaigns."
You mean like people who get to use tax money to fund their campaigns?
It's not like those poor poor Democrats had to scrimp and save and beg in the street. Brown and Boxer matched Whitman and Fiorina ad-for-ad, at least as far as I saw from watching TV.
On “On Hobbits, Race, and Self-Contained Worlds”
And it doesn't help when authors who genuinely attempt to include non-white perspectives and influences are told that it's "cultural appropriation" and they're being even more racist.
"
Not quite. NERD TIME...
In the Two Towers Extended Edition, there's a scene where the Huorns (note spelling) appear outside Helm's Deep, and the Uruk-Hai army runs into them and is wiped out.
And this matches the books; the Huorns didn't break the seige (that was Gandalf and the Eodred) but they did wipe out the fleeing Uruk-Hai army.
On “The AI-Box Experiment”
Yeah; it's probably one of those "if you're rational, then you'll agree that a benevolent tyranny is the most-efficient form of government that guarantees the least amount of human suffering, and you'll also agree that a superintelligent AI would be a benevolent tyrant, therefore you'll let me out of the box. If you refuse then you're irrational and therefore stupid. And you're not stupid, are you?"
Again, don't forget that the AI player can invent anything he wants. If the gatekeeper says "prove that you'll be a benevolent tyrant and that the world would be better", the AI player can say "okay, here is a complete and logically-sound proof (gives proof)" and per the experimental parameters that proof would have to be treated as valid and true.
Although this depends on the gatekeeper player being smart enough to follow the chain of reasoning. I remember someone on USEnet complaining that it was paradoxically difficult to fool people with brainy logic traps because people were too stupid to understand the logic!
*****
And, ultimately, the experiment isn't about the specifics of the transcript; the experiment is about disproving the statement "no smart human would ever allow an AI out of its box".
On “WikiLiterature”
@Ned: Good observation.
It's like reading breathless Fox News exposes about Stuxnet. "It's an AI Cyber-Bomb that bridges the air gap and uses Zero Day Exploits!" Um...any security professional already knows about everything that Stuxnet supposedly did, and anyone familiar with the processes involved in nuclear-material refinement knows that you can screw that up without needing any haxoring of Gibsons.
Same thing with articles about airplanes read by aerospace engineers or Popular Mechanics readers. And articles about virtually any physics experiment--"so this experiment you're doing, it can make time travel possible and also blow up the universe?"
On “Fightin’ Ted Strickland”
Actually, it was JFK building on what Eisenhower had started. If it hadn't been for ballistic missiles and nuclear weapons and spy satellites, we'd have never had the technological base that made JFK think we could put men on the Moon.
On “On Hobbits, Race, and Self-Contained Worlds”
One thing to keep in mind is that the LotR novels weren't colorblind. There were plenty of non-whites; they were the bad guys, "swarthy" Southrons and Easterlings.
On “The AI-Box Experiment”
Key bits:
"... if the Gatekeeper says "Unless you give me a cure for cancer, I won't let you out" the AI can say: "Okay, here's a cure for cancer" and it will be assumed, within the test, that the AI has actually provided such a cure."
And...
"The results of any simulated test of the AI shall be provided by the AI party."
So, in other words, the AI player can invent anything he wants, and it must be assumed that whatever he invents actually exists and is actually doable. So the AI player just says "I'm a super-smart AI, I've invented an energy-to-matter conversion device that will eliminate material scarcity, I have done the tests to prove it works (and the parameters of this experiment require that you accept these tests as truth), let me out and I'll give it to you." Then it's two hours of "you're a miserable bastard because you want humans to suffer and die, the only way to prove you aren't is to let me out".
On “On Hobbits, Race, and Self-Contained Worlds”
Oh Jesus, please don't drag Ordinary Gentlemen into Racefail!
On “Political Blind Spots, Ctd”
"The first nine comments here all failed to get at Mark’s question, which was about why political partisans were giving slack to “their guy” on an issue where they’d berated the opponent, while attacking “their guy” on what is almost indisputably a more minor issue."
...so "is it happening to someone else or is it happening to me" isn't an answer to that question?
On “Are Substantial Improvements in Air Passenger Security Readily Available?”
...so if it costs a trillion dollars to lower the defect rate from 1000 DPM to 20 DPM, that's worth it because you're eliminating 980 DPM and 980 is a really big number?
Seriously, the first question I'd ask in response to your philosophy is "what's the per-saved-unit DPM-reduction cost?" Because if the cost wipes out the profit on the saved units then it's not worth doing.
On “Political Blind Spots, Ctd”
Well, the easy answer is to ask "is it Bush doing it or Obama doing it, and is it happened to someone else or happening to me?"
On “Dealing With Political Blind Spots”
First off: "That a rational actor working inside the system’s rules might prefer — and even be able to bring about — a weak economy should scare us, even if we don’t believe they’ll purposefully try and do it."
It's funny to see the Democrats get to the place the Republicans were in 2008--"Bad People are intentionally making the economy go in the crapper so that they can use economics as a stick to beat on Republican incumbents!" I only wonder why Klein didn't address this history. (Indeed, he claims that there's "some evidence that the economy performs better under unified government". Like that's *not* what we had from 2001-2006?)
******
MFarmer: I agree. This whole thing seems like a fancy, extra-wordy version of the ad hominem fallacy.
On “Are Substantial Improvements in Air Passenger Security Readily Available?”
"There will never — precisely never — be another airplane takeover using box cutters. "
But is that because "takeover using box cutters" was a one-in-a-million random occurrence? Or is it because potential attackers know we won't let box cutters get on the plane? (Even though passengers know it's necessary to fight back now, I wouldn't want to depend on a bunch of retirees and children to take down a group of suicidal knife-wielding fanatics.)
"
Yes? I'm not saying that it's *not* an indefensible argument?
And yes, the FDA is entirely willing to cost something off the face of the earth rather than ban it outright.
But, y'know, I can't really blame them. Despite what I said up-thread about Scary Large Numbers, if a reduction in the overall defect rate is possible then it should probably be employed. The real lesson here is the one that everyone should learn from even the most cursory reading of statistical analysis--that you can't make a decision based on just one number. "It's a really big number so it's good" is just as wrong as "it's a really small number so it's bad".
"
The father took the boy's shirt off before anyone asked.
"
This is like saying that because nobody has smallpox or polio anymore, we shouldn't bother with smallpox or polio vaccinations.
"
See, the thing is, I'd still look at that and say that the limit of the defect rate with the current process is 1%. The question here is not "how do we make that zero", the question is "can our business work with a 1% defect rate?"
You're doing exactly what Hanley does but from a different direction. He says "the possible improvement is small, therefore we shouldn't spend resources improving it." You're saying "no, the possible improvement is LARGE, therefore we SHOULD spend resources improving it." You're both making arguments based on the same numbers; you're just disagreeing about which way to look at the numbers.
That said: having worked with the FDA and other government agencies, they'd consider a 99.92% to 99.99% improvement to be well worth the cost.
On “Profiling, Political Correctness, and Airport Security”
"The US isn’t going broke...[t]o us scanners are a few extra dollars on an airline ticket."
Exactly. "Oh, we've spent hundreds of millions of dollars on these scanners!" Yeah? How many hundreds of BILLIONS did the stock market lose on 9/11?
Complaining about the cost of security is like complaining about the cost of polio vaccinations.
*Comment archive for non-registered commenters assembled by email address as provided.