Facebook is scanning over 100 million photos per day, as its users upload photos and “tag” (or associate names with the people depicted) those photos. Computers are refining algorithms and learning your face. Automatic face recognition is becoming advanced and Facebook is making it a larger part of the Facebook experience. Imagine taking a photo of a STRANGER and learning his/her name, age and address within SECONDS. It’s coming… as are WILD LEGAL ISSUES. As always, it is fine to embrace technology, but always be smart and cautious of issues that could arise from it.
The warning first came from Facebook’s rival, Google. Eric Schmidt served as Google’s CEO from 2001 until recently. He claims Google created “face recognition” technology and stopped progressing with it because of concerns. Why? He said Google wanted to be a “DO NO EVIL” company and considers the moral impact of its inventions. He described the facial recognition software as “creepy” and “potentially harmful.” Wow, strong words, especially from a company that has a car running around nearly every city in the United States filming “street views” of exactly what our homes look like.
“If you imagine, for example, what a perfectly executing evil dictator would do with all this technology — complete supervision, complete tracking, and so forth — and then you imagine what the dissident in that society would do, using the very best encryption tools and so forth, unfortunately you conclude that exactly the same tools are the ones that would be used by terrorists against an open society.”
He also suggested some company will “cross that line.” Clearly, he knew that company would be Google’s rival, Facebook, and the internet giant has come close to doing just that. Of course, you have to look at some of the dramatics as “business puffery.” However, Facebook’s version of facial recognition software has examined over 23 Billion photos over the past few years. It continues to review billions and Facebook is allowed to collect information per its user disclaimers- even metadata- the hidden stuff like GPS and dates in your photos.
Facebook has and will continuously advance its “Tag Suggestions” features, using facial recognition technology to speed up the process of labeling friends and acquaintances that appear in photos posted on Facebook. This is made easier as the computer database “learns” the faces from the billions of photos in its system. However, experts say that faces are not so different and technology is not yet at the point where it can pick out faces, with consistency, in more than a small ratio. In other words, don’t expect Facebook or any other company to be able to automatically associate your face with your name outside of comparing it to just a few hundred others. But that is the next step.
The technology has raised eyebrows in some circles. Some allege Facebook is invading rights with its failures to disclose that it is (and has been) using face recognition software to advance “tagging” to the next level, which for some will make sharing photos very simply and almost immediate. Facebook has countered these claims noting that the photo-tagging suggestions are only made to newly added photos, that only friends are suggested and that users can disable the feature in their privacy settings. Right now, it is harmless enough and a pretty exceptional technology… unless you think about it at the level Eric Schmidt thought about it.
500 to 800 million users, the ability to collect data on each and every member, the ability to “map” each user’s face (as well as the faces of non-Facebook friends) and geographically sort these based on location of upload or other tags and, well, you begin to raise privacy issues.
There is little law on the books or on the horizon. Eric Schmidt says he expects LAWS WILL FOLLOW the technology. He urged lawmakers to hold off and see how advances developed instead of preemptively overreacting and stifling technology. We agree, but there is grave need for monitoring and legislating to protect abusive use of this data.
Add just a little advancement in the technology or rogue software using it and one could easily see a situation where Facebook, or facial recognition, begins to intercede with reality. How? A guy sees a cute girl on the street and snaps a quick photo of her on his smart phone. Within seconds, the algorithm gives him a few local choices, he picks the best match and has her name and other information right at hand- creepy and an invasion of privacy. What if a criminal noticed her in a store buying a nice item or withdrawing money at a bank? That same photo could lead to information that could lead to the opening up of a network of information whereby the crook could click his way all the way to her home address. It’s not that unrealistic and is the next step of concerns. It ends with a system whereby we all can be digitally identified more easily and privacy further fades into the past.
I criticize Facebook’s decision to automatically enable the facial-recognition technology for Facebook users, rather than allow them to OPT-IN. Further, Facebook needs to protect and make sure no information is available to the public from its computer database and that other applications and software cannot tap into it. In fact, it probably shouldn’t be “studying” faces of those who have opted out.
Check your privacy settings if you want to opt-out of this feature. You do so at the screen above. Stay tuned for more updates as we learn more about this interesting issue.