IE 11 is not supported. For an optimal experience visit our site on another browser.

How Microsoft is helping Facebook fight child porn

Facebook takes child pornography very seriously — as we can see by recent crackdowns — and now the social network is gaining a new weapon in its battle against one of the Internet's vilest by-products. A Microsoft-powered weapon, that is.The New York Times reports that Facebook is now using PhotoDNA, an image-analysis technology developed by Microsoft, to combat child pornography on its pa

Facebook takes child pornography very seriously — as we can see by recent crackdowns — and now the social network is gaining a new weapon in its battle against one of the Internet's vilest by-products. A Microsoft-powered weapon, that is.

The New York Times reports that Facebook is now using PhotoDNA, an image-analysis technology developed by Microsoft, to combat child pornography on its pages. (Msnbc.com is a joint venture of Microsoft and NBC Universal.) 

PhotoDNA has been around for a while — in fact, Microsoft made it free and donated it to the National Center for Missing & Exploited Children back in 2009 — but it has been fine-tuned quite a bit since the first time we heard about it. And now Facebook will be the first major service to take advantage of it.

Until recently, the social network has relied "primarily on abuse reports from its users, reviewed by trained employees, to find and eliminate offensive images." With Microsoft's technology though, it should be able to prevent a lot of these images from ever even appearing on the site. 

This will be possible because PhotoDNA can automatically compare uploaded photos against those in a database of "about 10,000 images collected by the National Center for Missing & Exploited Children." (The center has a "Congressional mandate to act as a clearinghouse for this material, to help identify and aid victims and to assist law enforcement in investigations of perpetrators.")

The way the technology makes these comparisons is quite clever:

PhotoDNA works by creating a “hash,” or digital code, to represent a given image and find instances of it within large data sets, much as antivirus software does for malicious programs. However, PhotoDNA’s “robust hashes” are able to find images even if they have been altered significantly. Tests on Microsoft properties showed it accurately identifies images 99.7 percent of the time and sets off a false alarm only once in every 2 billion images, and most of them point to nearly identical images, [Dr. Hany Farid, a Dartmouth computer science professor and expert in digital imagery who worked with Microsoft to hone the technology] said.

To create a hash, the software puts the image in black and white and into a standard size. Then it carves the image into blocks and subjects it to an array of measurements. The resulting “signatures” can be provided to online service providers, who can then use them to find these specific illegal images on their systems without possessing them or looking at customers’ private content.

It sounds like PhotoDNA could have a significant effect on Facebook's efforts to track down and eliminate any of the child pornography which could slip by in the 200 million images uploaded by its users each day and Microsoft hopes that major Internet services other than the social network will seek to adopt the technology as well.

Related stories:

Rosa Golijan writes about tech here and there. She's obsessed with Twitter and loves to be loved on Facebook.