Slashdot Mirror
Face Search Engine Raises Privacy Concerns
holy_calamity writes "Startup Polar Rose is in the news today after announcing it will soon launch a service that uses facial recognition software, along with collaborative input, to identify and find people in photos online. But such technology has serious implications for privacy, according to two UK civil liberties groups. Will people be so keen to put their lives on Flickr once anyone from ID thieves to governments can find out their name, and who they associate with?"
ANYTHING you do online is NOT private! PERIOD!
Polar Rose relies on a combination of our unique face recognition algorithms and the collective intelligence of our users.
They seem to have made a fatal assumption.
I wonder if I use bold in my signature, people will notice my posts.
Who says they aren't already doing this? Unlike your credit report, you can't see everything they've been gathering on you.
I do some street photography and, although I don't personally publish material on the web, some of the people who hire me do. So even if you don't put your photo on Flickr because you are afraid of being identified by search engine there is nothing stopping me from putting it up there for you.
"Sacrifice for the good of The State" - The State
Startup Polar Rose is in the news today after announcing it will soon launch a service that uses facial recognition software,
The only "facial" recognition software I use is Google Image Search with Safesearch turned off.
The theory of relativity doesn't work right in Arkansas.
let's face it -your information is out there somewhere. Instead of being afraid of getting involved in some online community, let's think of better measures of protection against identity theft.
Unless face recognition has improved drastically, this company will just fail like the last couple companies which attempted to do anything with it.
Ach, so I will finally be able to look up this lovely Russian girl I've met online.
She's been sending me pictures of herself (chuckles) and her name is Sonya..
She's SO sexy she's got me worried, but my worries will finally go away as soon as I check her photo with this new service!!!
I agree with the parent. Anybody that posts photos of him/herself on the net should reasonably expect that anybody will see them. This is the reason that I am a bit uncomfortable posting my bookmarks to del.icio.us.
My advice to anybody who wants their cake and eat it too: Use different handles for different applications.
That is, if you want to indulge in the MySpace/LJ/VOX blogging, then use a handle unique to that type of activity (eg. BlogUser99).
If you want to indulge in Flickr/Photobucket/Picasa photo-sharing, then use a different handle (eg. PhotoDad12).
The same goes for social bookmarking and product reviews on Amazon and the like.
And, of course you should never use your full name except for in business transactions.
By using different handles, you'll give black hats/feds/5kr1p7-k1dd13z a hard time trying to figure out who you really are.
Just my 0b00000010..
idm owns me
I'm curious how things like this will work with Witness Protection.
Setting aside the fact that, at least right now, sunglasses fool these systems... if someone, lets say, a member of the Talini Crime family wants to find a rat. By giving a picture of him to this company, they could then search for pictures on the internet he appears in.
Considering how many pictures people take with random people in the background, it seems inevitable that said rat would turn up.
I agree with the parent. Anybody that posts photos of him/herself on the net should reasonably expect that anybody will see them.
What if someone else took a picture of you, or say, your wife, or kids, and put it on the net without your consent? Would that be ok? It's not always about what you would do with photos of yourself, but what other people do with your image that you have no control over.
Not a big deal, unless you happen to work for a conservative company and maintain an anti-government blog or some such thing.
"Si vis pacem para bellum" -Publius Flavius Vegetius Renatus
It will not be very long (a decade? two decades?) before face, body, gait, license plate, voice, speech, handwriting, textual habits, (and so on) recognition software will be powerful enough to recognize people in real-time, from a variety of real-time inputs.
Even the past will be open to analysis, a theme called "retroactive surveillance." For example, the Seattle bus system keeps timestamped footage of people coming in and out of the bus, and the Seattle bus system keeps records of where the buses are, and when, by GPS. In theory, these two systems can be correlated, and, if you have a system for analyzing faces, you should be able to connect the "network of data" to figure out who is where and when. This type of correlation is what software visionaries are working hard to achieve, with efforts such as the Semantic Web.
People who are worried about "the mark of the beast," through such things as RFID tags and so on, are worried about the wrong thing. You won't need to "wear" anything. You won't need any special marks, once software is sufficiently capable. Your face, your clothes, the way you walk, your posture, the regular patterns you follow every day, your voice, all are sufficient enough, in themselves, to serve as the "mark of the beast."
It is conceivable that you will be able to limit government use of this sort of technology. But will you be able to stop private users from using this sort of technology? If you envision a future revolution of some sort, do you believe that the revolutionaries would not use this technology themselves? To track the motion of police vehicles, and individual policemen, and the people who work for and against you?
The underlying activities behind these technologies: Collecting information, seeing, hearing, sensing, and then correlating what is seen, what is heard- these are foundational. The "problem" is simply intelligence, itself.
I doubt that willful blindness or doubt is going to help us in our path to the future. We see that backwards countries practicing willful blindness, not advanced ones.
For example, don't say require FISA warrants before listening, because it is quicker to buy another disposable cell phone than it is to obtain a FISA warrant.
They have NEVER needed FISA warrants before listening. In the event that they need to tap in an emergency (where waiting for a FISA warrant could lose the chance at intelligence), they can just start doing so. However, they do need to apply for a warrant within 72 hours of starting the tap. How could any reasonable person have a problem with this? All they are saying is that you cannot wiretap without ever telling anyone about it.
We obviously need some sort of security. What is OK?
Yes, we do. But we cannot forget that we have a system of checks and balances. Democracy does not move as fast as a dictatorship, and a dictatorship can (in theory) move much faster to protect its citizens. If that is what we truly want as a country then let's just do it and quit pretending. This whole "we are still a democracy with governmental checks and balances but because the president declared war on an abstract concept he can do anything he wants" thing we have going now just does not make sense. The excuses are always so flimsy, it is always a claim that it is perfectly legal under written law and when that proves to be false then it becomes "oh well, none of that matter anyway because he's got unlimited wartime powers".
But you ask what we can do? Obviously we are doing some things that make a lot of sense. Better information between the intelligence agencies is a no-brainer, and I would go as far as saying going after the Taliban in Afghanistan was a good move as well (Iraq was obviously a horribly stupid blunder/distraction though).
However we do a lot of stupid things also. Hiring a lot of poorly trained rent-a-cops to play detective in the airports was probably not the best use of our resources. Insane restrictions on what we can take on airplanes do nothing for security, but make ignorant people feel safer. The whole slew of ways we try to throw billions in poorly thought out "technical" solutions like RealID, MagicLantern, facial recognition (which doesn't work any better than space lasers shooting at ICBMs), and whatever kludged algorithm generates the Mo-Fly list do nothing for security and cost both money and civil liberties.
There are many tried and true intelligence gathering and counter-terrorism techniques, but the current administration is more interested in presenting a color coded security theater for the masses complete with high tech sounding ("it involves computers so you know it must be good") projects. The paranoid thinks they are just using "terrorism" as a bogeyman to implement systems to track and control all citizens. I actually think that is a side effect of their actual motivation to dump money into their friend's and contributer's companies.
Finkployd
As other posters have pointed out, once something is online, it's not private anymore. Complaining about the 'privacy concerns' of this software bugs me, because it's a distraction from real privacy issues.
Reminds of the Libertarian Party (of which I am, unhappily, a member) - seriously complaining about trivial issues means that people will trivialize your complaints about serious issues.
In a technical (and technological) sense, you're absolutely right. Given the nature of digital information, anybody putting any information online would be well advised to act as though it is going to get back to everybody they know, perhaps through channels that don't even exist at the time you put the information online.
But the more complicated social reality is that in most people's experience, the public-private distinction has usually been one of probability and degrees, not an all-or-nothing proposition. It used to be the case (and still is, though less and less so) that you could go to certain technically public places and still have a practical/probabilistic expectation of privacy. For example, you could go to a political or cultural event for an unpopular group (a gay pride parade, for example) and have a reasonable hope that it wouldn't get back to your employer or family. You might be in a technically public space and you (hopefully) knew you were taking a risk, but the risk was small enough that it was worth it.
The problem raised by this kind of technology is that it is eliminating those kinds of physical and virtual spaces -- the spaces where you can meet and interact with others and have some practical (if not airtight) expectation of privacy. The fact is, there are very few real places you can socialize with lots of other people that have a truly complete expectation of privacy, so the probabilistic expectation is often the best you can hope for. For people with some kind of politically or culturally marginalized interest -- and let's face it, who doesn't have at least one interest that falls into that category -- it's a sad development.
The fatal flaw in this process, is that it's Tom Cruise's face. Can't I have one day where I don't have to look at that guy?
This is why I worry about this process. Not because of my privacy, but because there are already too many instances of my large face on the Net. You people have enough of me to deal with already, without yet another database crunching my oily pixels and spitting them back up at you with a hyperlink attached.
Slashdot Burying Stories About Slashdot Media Owned
Are we really going to pretend we didn't know technology like this was coming? Are we going to act all heated with righteous indignation about something that researchers have been chasing after for decades?
Everybody knew about it and expected this technology to be perfected sooner or later (and for now it seems that it's still a bit later). So, if you were that worried about someone being able to Flickr and Google your personal relationships together, you should have thought twice about putting your entire life up for digital scrutiny in the first place.
The privacy problem isn't with this technology, it's with people who put their personal life on the biggest computer network ever, freely accessible to all and then expect it to be private.
They need to get their head examined and by the looks of it, that's exactly what they'll get.
I think many of the people submitting their images to dating or adult sites should start worrying right now...
Agreed. I submitted a story to
http://www.msnbc.msn.com/id/15994151/site/newswee
From that story, a good example:
Cheap video technology (esp. video-capable cellphones) and social sites make it all possible.
Simply being in public can get you on these social sites, whether you actually use them (or have even HEARD of them) or not. In the end, the only way to ensure your privacy is to not become a part of society. If you venture into public, you too could end up on some social web site.
And remember--this is the PUBLIC engaging in a type of surveillance on the PUBLIC. For the tinfoil hats out there, it's not just the government's watchful eye you have to be careful around; it's that video-capable cellphone in the hands of the seemingly innocent rider sitting across from you on the train, too.
Government's idea of a balanced budget: take money from the right pocket to balance...oh who am I kidding?
>For example, you could go to a political or cultural event for an unpopular group
Parent makes an interesting point. Who would risk going to any public protest for anything (war, whatever) knowing that you will probably turn up in a Google image search for doing so?
Steve
A work that expires before its copyright never enters the public domain and thus enjoys eternal copyright protection.
A friend of mine works for a security firm here in NYC. They do camera system installs for certain *really*, *really* high security locations. If he wanders around in certain areas of the city, he'll have a nice email the next morning retracing his steps with still photos at various locations. The surveillance operators just feed the system a headshot and the rest is history. Sure, it's a little joke amongst co-workers, but it's fully possible today, right now.
Ok well to start, you cant fight a philosophy. What you are perhaps asking is that the government do more to fight Criminals. I would say that we have enough laws already to fight criminals. Some would say that this is a brave new world we live in and they need better tools to keep up with the crimes. I disagree that the world has magically changed and that we need to become a police state to fight for security. You will never be secure, because security is a concept of the mind (thats why gun enthusiasts think that they can buy security, they are somewhat right). Even your american founding fathers knew this in what was argueably a more savage and brutal world than the one we live in. They say clearly, dont sacrifice liberty for security, and I think whatever country you live in that thats a good idea. I would suggest that you instead look at the root causes and motivations of these particular criminals. Bin laden has said specifically what he wanted, most notably the USA out of the middle east. Why not start with that?
I'll just use my special getting high powers one more time...
The issue I have with this is, again, the disposable phone. Habib Mohamed buys a disposable phone to call his mum, Bosama InLaden in Pakistan. The Feds freak and immediately start listening in, only to find out that Mr. Mohamed really is calling his mum and stop listening and remove Mr. Mohamed from any further surveillance. The problem is, they would not have known this if they had not listened. And they really can't go to the FISA court now to get a warrant for tapping Mr. Mohamed's calls to his Mum.
Why can't they? They go to the court and say: "Based upon this evidence which shows a legitimate need to wiretap Mr Mohamed's phone ASAP we started tapping, attached is the evidence and our application for a warrant". Then after tapping him for a while (assuming the warrant is issued, which it almost always is anyway), if they determine that they were wrong and they are getting no evidence, they stop the tap and everyone is happy and legal.
There is no "punishment" if no evidence is gathered in a tap, BUT they have to have a solid case for one when they apply. If they do an "emergency tap" to get a specific conversation they feel will result in critical intelligence, but none comes up, they STILL have to submit the paperwork to the FISA court and explain why they felt they needed to tap.
The purpose of this is twofold. The court will obviously not allow tapping for political gain (imagine if Nixon tried to get federal judge approval for Watergate). But most importantly, it holds the Executive branch accountable for their actions. Without having to apply for a warrant they can tap whoever they want, whenever they want, on any whim. Fishing expeditions, political gain, corporate spying for their buddies, why not? Nobody except for them will ever know. Even if you do trust this administration, are you prepared to afford this level supreme trust to every future administration?
Finkployd
Maybe a robots.txt -esque way of opting out of your picture's being indexed?
The more important question is, why not have oversight in hindsight? I can think of only two reasons: 1) The powers that be don't want any records of their wiretaps that they themselves cannot hide/destroy/deny. 2) The powers that be want to continue with long term wiretaps that they know would never hold up in court. Neither is particularly palatable to me.
An interesting anagram of "BANACH TARSKI" is "BANACH TARSKI BANACH TARSKI"
It's odd, everyone seems to be fixated on the fact that people could use this to find images of them. The trouble I see with that way of thinking is that if an image you don't want viewed is on the internet it is there for one of 2 reasons
1)You put it there. In this case it's your own fault and you shouldn't complain.
2)Someone put it there without your permission. Think naughty landlords with hidden cameras or stalkers with telephoto lenses. In this case you generally don't know you're on the $/month "gentleman's website",interent shrine of undying love or heck, even some jerrk at work's myspace page and this image searcher has the possibility to point the fact out to you before someone you know spots it.
I would think people that don't want to be exposed on the internet would be happy to see something like this come out so they can see just how exposed they currently are.
An open source version of a content-based image search engine is already available in the Carnegie Mellon-Intel joint research project called Diamond. (http://diamond.cs.cmu.edu/) It has a facial recognition application called SnapFind which can already perform arbitrary face matching and other filtering (given the image data), and it is available for free. Disclaimer: I work on Diamond, which has many other applications than this, and thought it was appropriate to post.