“The statements related to Google are completely untrue.” That’s Google’s response to the assertions from this story, which I posted about here.
Update: Sorry folks, to clarify: This was an official Google spokesperson’ statement.
Google On CIA: Untrue
"The statements related to Google are completely untrue." That's Google's response to the assertions from this story, which I posted about here. Update: Sorry folks, to clarify: This was an official Google spokesperson' statement….
I can only stand by my own words, and report the words of others. Robert Steele, for his part, is standing by his statements.
That statement could mean they are helping, just not specifically as described. If they really were not helping one would think they’d say “We do nothing to assist the Government with information gathering”. Frankly, I think they should be working in various ways to help the government curb potential terror attacks.
“The statements related to Google are completely untrue.”
Who said this at Google? Is this an official, public statement? Or is it a private, insider reaction?
Google has developed some of the most sophisticated searching technology in the world. It would be a disgrace if the government didn’t try to work with them. And just because the two organizations might be working together, it doesn’t mean Google can’t sell technology while still “vigorously protecting its users’ privacy”. Google already sells Google Search Appliances to corporations and organizations (like the World Bank). Why shouldn’t the government leverage the same technology?
Jarid: You’re missing the point. Of course Google should be able to sell Search Appliances to the government.
The issue is whether Google is sharing information on what each of us are searching, in order to identify patterns and trends in our search histories, in order to determine which of us are terrorists.
For example, the following story was told to me back in 2003. It involves a “friend of a friend”, so take it with a grain of salt. The number of hops is small, but my caveat is that it is still something that I do not know directly.
That said, I heard about a fellow who, in 2002, got quite interested in all things “terrorist”. As a history buff, not as a participant 🙂 He wanted to find as much information as possible, read as many things as possible, understand as many things as possible.
So he began using Google and searching for all sorts of terrorist-related words. He started with the obvious ones.. Osama, Al Quaeda, etc. Then, when he would discover new information on some web page, he would use that new information as search keywords to try and find even more information.
Well, after about three solid weeks of doing this, I am told that two men in suits showed up at his front door, and wanted to interview him. They didn’t take long, and left shortly thereafter, apparently satisfied that this fellow was a curious history buff, not a terrorist. So all in all, as far as I can tell, no harm done.
But the question still remains: How did these men know to show up at his house? I can think of two main possible explanations: (1) The intelligence agencies are tapping every single ISP and every single IP endpoint, looking for suspicious activity. Or (2) the intelligence agencies have intelligently placed a select few information gathering taps at key points, i.e. at a place where almost everyone goes to actually search for information.
I am a scientist, so the scientist in me wants to believe the occam’s razor, simplest explanation. Remember, this fellow used Google, exclusively. What better way to identify suspicious activity than to work with the company actually gathering all this information, rather than trying to individually tap every single internet “tube”? Which approach would you take, if you were an intelligence agent?
So, who really knows for sure? Not me, I guarantee that. But let’s assume for a moment that Google is not sharing any information with government agencies, under any sort of non-disclosing, secret court order. Why doesn’t Google then step out and try to protect our privacy a bit more? In other words, why doesn’t Google use some of that wonderful Ajax to encrypt our queries and decrypt the results on the client side? Won’t that help keep the terrorists, hackers, and other evil-doers from snooping all our data, and planning further attacks against us? Why, if there is a risk to our privacy, isn’t there more being done about this?
Sure, It’s not CIA – but NSA that works with Google.
Ha! Men in suits showing up after Google searches for certain keywords relating to jihad? Hardly. Too obvious. That’s such an easy bait for the EFF or other journalists to drop. How simple to measure overt government surveillance and ruin a billion dollar corporation’s credibility. Oh you don’t believe so? Then do it yourself, search for those terms repeatedly and see what happens. Nothing.
A more realist collusion between a domestic intelligence service and a patriotic American search engine is a high ranking SERP. The CIA could (nay should!) have its own clandestine websites featuring content that contains enemy propaganda. Google would graciously provide that site as a number one result under searches for certain keywords that would interest an enemy combatant. (Nothing too obvious like `jihad` or `Amerikkka` or other such nonesense, but certain other phrases which are unique to enemy propaganda in other languages.) The CIA would monitor and tabulate access to their covertly operated sites, especially accesses referred from Google, and then cross-reference what is known about the originating IPs at foreign telecoms. A tactic like that is…just…duh!
In fact, if the CIA was really sophisticated, they would have dozens of unlinked(!) “content honeypots” on a variety of `borderline` topics (insurgent ideology, transportation defense, mass media psychology) and review user access at these sites against each other. Why?
Build your own web of content. Make the database of intentions work for you. Tempt others with know-how and deduce their motivations. You can monitor them and no one will ever be the wiser. As Google webmaster guidelines say “Webmasters who spend their energies upholding the spirit of the basic principles [of quality] will…enjoy better ranking” Hell, you may not even need to collude with Google to achieve that. You can have “persons of interest” sniffing your site if you just offer them something they finding “interesting.”
Damn. This deserves its own blog post. Guess I’ll start writing…
Raffy writes: Ha! Men in suits showing up after Google searches for certain keywords relating to jihad? Hardly. Too obvious.
Look, I disclaimed that I do not have first-hand knowledge of this. The account is from a friend, who swears it happened to his friend. Even still, like I said, take my story with a grain of salt.
But go back and read a little more closely what I said. I didn’t say that this guy just typed in “death to Amerikka” or “what does Osama want me to do next”. I said that he started with obvious keywords, then used non-obvious, more rare keywords that he would find on the initial results pages as his next queries. He kept a list of new, interesting, unique keywords, and continued trying those, as the weeks went by. He was actively seeking out those non-obvious keywords that would be needed to find the pages that one would not otherwise find.
Now, look at what you propose: Google would graciously provide that site as a number one result under searches for certain keywords that would interest an enemy combatant. (Nothing too obvious like `jihad` or `Amerikkka` or other such nonesense, but certain other phrases which are unique to enemy propaganda in other languages.)
Well, guess what? If this friend-of-a-friend of mine was acting in exactly the manner I said, he would eventually start finding all those GoogleRank#1 pages you are talking about, because he would be using all those phrases and keywords that would interest an enemy combatant.
In other words, this dude would fall quite cleanly into all these honeypots. Because this dude was actively trying to seek out all those sites that an “enemy combatant” might be reading. Like I said, he was trying to find out as much as he could about the whole terrorist thing, on the web. From the other side of the computer screen, he would look just like a “lost” sleeper trying to reconnect with his “unit”.
So who knows if it really happened or not. I’m just passing along the gossip and hearsay that I have heard.
But Google just officially announced that they do not collude with the intelligence communities. So if that’s true, and if my story is also true (lots of hypotheticals here), one really has to wonder how all this intelligence data is being gathered without using such an obvious central repository like Google (or Yahoo or MSN or Ask or all of the above).
JG, no value judgement on you personally here. I am skeptical of the veracity of your hearsay – which you, yourself adequately disclaim. I simply point out factors I feel are obvious that make the story not very plausible. As I said before, anyone could simply keep digging for keywords upon keywords and wait for the men in black to show up. I assume people are doing this right now. But it simply won’t happen. Too obvious. Too easy for the CIA (or whomever) to get caught in a PR nightmare.
You wonder “how this intelligence data is being gathered without an obvious central repository” but that’s exactly what I describe. “Content honeypots” as I coined, preclude the need for search engines to turn over their records so blatantly. It is possible there is covert monitoring going on at a number of websites that intelligence agencies front. These agencies will never *act on* suspicious accesses by showing up at your door. (It may prompt them for continued surveillance on a suspect if undetectable, but you need to do something blatant before they will give themselves away and question you.)
So in summary, Google doesn’t need to turn anything over to the government. They just make the CIA’s job easier by including their fake websites in SERPs. That’s my theory. If the men in black show up on your door to question anyone reading the propoganda, well then they just gave everything away – blew their own cover. It wouldn’t be hard to find the exact site which alerted them by process of elimination either.
Raffy: Thank you; I am not reading your words as any sort of personal value judgement.
You have a valid point. I think I can agree with you that Google need not share everyone’s queries in order for this scenario to proceed.
But look at what it takes in order for this honeypot scheme to work. It takes Google colluding with the intelligence guys to place these honeypots at the top of ranked lists. And Google has also publicly stated that it does not hard code. So now this appears not to be true. One way or another, something doesn’t quite add up.
Oh, and one more relatively minor point. You write: If the men in black show up on your door to question anyone reading the propoganda, well then they just gave everything away – blew their own cover.
Well, I hate to break it to you, but there have been actual cases of intelligence guys actually showing up just like this. Back in 2002, also this same time period as the other “friend of a friend” I heard about, a Massachusetts professor (wish I could find an earlier link to the story; sorry!) was falsely accused by a work colleague of being a terrorist. Instead of parking themselves secretly and watching for a few weeks, the FBI instead showed up on this guy’s doorstep and interviewed the guy for 20 minutes. And rapidly determined he was indeed not a terrorist. But they did “blow their cover” and show up, anyway.
So again, my story is still “friend of a friend”, but it does not seem that implausible. Especially if the intelligence guys really didn’t think he was a terrorist, but had to do their due diligence and check him out, anyway.
Anyway, your other points are quite valid.
JG, you said:
The issue is whether Google is sharing information on what each of us are searching, in order to identify patterns and trends in our search histories, in order to determine which of us are terrorists.
But, my point is that people were jumping to conclusions that just because the two might be working together, that in fact, it was around sharing user data. There’s nothing in the story that indicates that. Read the original story again. All it alleges is that Google provides “assistance” to the intelligence community by helping them leverage Google’s data mining capabilities.
Working with the government and protecting user data are not necessarily mutually exclusive.
John, is that the whole quote from the spokesperson at Google? I would like to see if we can politely pose a few specific questions publicly to google about this.
eg, “Has google released search-related information to any branch of the government, or government subcontractors?”
It just seems like their response comes too easy, because “the statements about google” are so broad.
I think a lot of us would be honored if John B would politely pose these sort of questions.
Raffy:
In fact, if the CIA was really sophisticated, they would have dozens of unlinked(!) “content honeypots” on a variety of `borderline` topics (insurgent ideology, transportation defense, mass media psychology) and review user access at these sites against each other. Why? Build your own web of content. Make the database of intentions work for you. Tempt others with know-how and deduce their motivations.
What you’ve just described is very like what one organization does in the Michael Chrichton novel, State of Fear. In that novel, the organization builds a network of sites with information, propaganda, etc, to tempt would-be eco-terrorists, so that the organization could keep a better eye on what they were interested in, and planning.
Like most Chrichton novels, there was a lot of techie-philosophy, in which the author described how everthing today is networks of things, how the bad guys have their sophisticated networks, and how the only way to join the bad guys in battle is to have your own sophisticated network.
That and a *ton* of information on climate change. I didn’t know California was covered in glaciers 20,000 years ago; guess I should’ve. But that’s another topic …
Frankly, if Google says that they are not working with the CIA, that might be the best proof around that they are indeed working with the CIA.
Just how they interact is the interesting question here.
Jarid: Yes, I might indeed have jumped to that conclusion, that Google/intelligence collusion translated directly to sharing user data. The scientist in me found it to be the simplest explanation to the “friend of a friend” story that happenen in 2002. But Raffy posited another possibility: Google link honeypot hard coding. But I don’t believe that either, because of Google’s strong statements against hard coding. That would be a huge PR nightmare if it came out that Google hard-coded certain queries, without telling any of us which ones. It stains the whole purity of their SERP algorithms.
So can you think of another way Google might actually work with the government, without violating privacy, and without violating any of its other publicly-stated principles, such as the hard-coding of honeypot terrorist-related search result links, as Raffy suggests? I cannot think of a single scenario in which Google cooperates or colludes, without violating one of its principles. Can you? Would you share it with all of us, so that we might understand a bit better, and feel better about the whole thing?
It would be useful to get specifics on who at Google denied this. I am quite positive that Google is taking money and direction from my old colleague Dr. Rick Steinheiser in the Office of Research and Development at CIA, and that Google has done at least one major prototype effort focused on foreign terrorists which produced largerly worthless data.
Hopefully Google learned from Bill Clinton that the denial is ofter more costly than the deed when it completely undermines ones integrity.
CIA is *not* very sophisticated. In 1986 they knew the 18 functionalities for an all-source analysis workstation (Google for CATALYST and CIA) and they *still* don’t have it. CIA is a kludge of contractor provided stovepipes, none of which play well together.
I like Google. I think they have enormous potential. I think they are seriously stupid to be playing with CIA, which cannot keep a secret and is more likely to waste time and money than actually produce anything useful.
Best wishes to all,
Robert Steele
ISP’s Pay Anticompetitive Bribes to Google
David Orrell of the UK ISP Hotchilli Communications let it slip today in a private coffee meeting that they, along with other major ISP hosting providers are actively involved in protecting their home domestic market share and extortionate pricing structures by privately paying Google in developing Google’s local region search criteria so that it actively identifies UK targeted websites that are hosted overseas and penalises them to the extent that they do not show in regional results! David suggests that Hotchilli’s Technical Director – Andrew Turner was pleased that “the result of Google’s new search methods are stopping website owners from looking offshore to better providers of a lot lower cost…” thus maintaining bumper profits for UK based TELCO’s and ISP’s. Should this practice be encouraged?
This is absolutely BOGUS! Google needs to release a full press statement to the affect of “We offer no goods or services to the Government that cannot be used by the general public”. I have no problem with Google working with the US government, so long as it is Not in SECRET!
Until then, Robert Steele’s words should be taken 100% seriously!!!
Hello, I know this is a bit of an old topic, but there are several things that I would like to point.
This is not strictly about the Google – CIA relationship, if any. This is about the Google – People – CIA relationship. I am not affiliated to either (except the people, of course) but I see both good things and potential bad things at stake here. The bad things apply if and only if the process isn’t fine tuned and, at the same time, receiving feedback from all the parties involved.
There are a lot of ways in which Google and the CIA could work together for the good of the people. Imagine a counterfeiting operation, for example. A location-based modeling of the places of reported counterfeit goods or finances could lead to a quicker apprehension of the bad guys and to less damage sustained by private institutions, therefore to more good for the people then if the counterfeiting process were to continue (if we presume that the state uses its non-counterfeited resources to support the people and to apprehend the bad guys, of course). Another point would be theft or loss of certain electronics. Items that have either a GPS or a way to link up to the net can be built to automatically let the authorities know where they are, via an automated email, if the devices is broken into, an email that might contain the fingerprint of the thief for example (if the touchscreen/pad is made to collect it). More good for the people (less items lost and stolen), implemented and data-crunched by Google (because they know their stuff when it comes to both soft and hard tools) enforced by the state (or the CIA, in major felony operations). These are only two examples of the mechanisms that should be allowed to work, in this triangular relationship.
Then there’s the matter of abuse. Let’s take online piracy, what can be set up to avoid abuse is a random, but anonymous and voluntary, Jury of 12 that can vote on wiretapping files to decide if they are worthy and ethical to pursue or not. In this scenario, the public could just relax while browsing the web. If that doesn’t work, someone could buy the file-sharing sites that are active and use the advertising income on the site to pay the content providers (for the ripped movies, music and software) according to the number of downloads per item. File-sharing sites can also be a novel way to advertise new content before its general launch to create hype in markets. There are a lot of mechanisms that one could think of to make sure that all three poles of this hypothetical relationship are happy together.
The thing to realize is that Google and the CIA are made of people, and that they can actually help people provided that they implement feedback cycles to their operating procedures that involve constant input from the general population, non-affiliated to neither Google and the CIA.
So, regarding this Robert Steele, pardon me but he somewhat sounds like a comic book character.
Google lies (and does great evil). The comments were made by the top industry analyst whom I have known for a decade. For those who continue to fail to understand that Google is living on investor cash, earning a dollar for every ten it spends, I recommend The Google Trilogy e-book series by Steve Arnold. I do not lie. Evidently Google does lie, but that should come as no surprise to anyone with a brain.