WASHINGTON: Facebook, Microsoft, Twitter and YouTube are joining forces to more quickly identify the worst terrorist propaganda and prevent it from spreading online.
The new program announced Monday would create a database of unique digital “fingerprints” to help automatically identify videos or images the companies could remove.
The move by the technology companies, which is expected to begin in early 2017, aims to assuage government concerns — and derail proposed new federal legislation — over social media content that is seen as increasingly driving terrorist recruitment and radicalization, while also balancing free-speech issues.
Technical details were being worked out, but Microsoft pioneered similar technology to detect, report and remove child pornography through such a database in 2009. Unlike those images, which are plainly illegal under U.S. law, questions about whether an image or video promotes terrorism can be more subjective, depending on national laws and the rules of a particular company’s service.
Social media has increasingly become a tool for recruiting and radicalization by the Islamic State group and others. Its use by terror groups and supporters has added to the threat from so-called lone-wolf attacks and decreased the time from “flash to bang” — or radicalization to violence — with little or no time for law enforcement to follow evidentiary trails before an attack.
Under the new partnership, the companies promised to share among themselves “the most extreme and egregious terrorist images and videos we have removed from our services — content most likely to violate all our respective companies’ content policies,” according to a joint announcement Monday.
When such content is shared internally, the other participating companies will be notified and can use the digital fingerprints to quickly identify the same content on their own services to judge whether it violates their rules. If so, companies can delete the material and possibly disable the account, as appropriate.
Most social media services explicitly do not allow content that supports violent action or illegal activities.
“We really are going after the most obvious serious content that is shared online — that is, the kind of recruitment videos and beheading videos more likely to be against all our content policies,” said Sally Aldous, a Facebook spokeswoman.
The White House praised the joint effort. “Today’s announcement is yet another example of tech communities taking action to prevent terrorists from using these platforms in ways their creators never intended,” said National Security Council spokesman Carl Woog.