【】

When you send a tweet, post a Facebook status, upload a video to YouTube, or do anything on the internet, there's always a chance that little act will ripple out onto other social media platforms, taking on a life beyond your control.

Extremist content -- whether that's Twitter accounts spreading Islamic State or white supremacist propaganda, beheading videos posted to YouTube (which is owned by Google), or Facebook posts filled with hate speech -- is no different, which is why Facebook, Twitter, Google and Microsoft are putting together a database of "terrorist content" to better track and eliminate it.

The companies announced the development of a combined database earlier this week, an effort to identify and stop images, videos and other posts that can easily hop to one network when they're banned on another. They'll use "digital fingerprints" to mark anything that one company considers extremist, though each company will decide on its own whether the video or image or post will be allowed on its platform.

This should allow each company to more efficiently mark and take down anything they think is "terrorist content" that violates their individual policies. If Facebook has never seen a particular image, it could be hard for the company to immediately recognize it as something that violates their usage policies. But if, say, Twitter has already seen the image and identified it as, say, ISIS messaging, then Facebook employees will know to quickly take a look at the image.

The specifics of how all this works, though, are murky, making it difficult to know what kind of videos, posts and images the companies are evaluating. Twitter, Google and Microsoft did not respond to requests for comment, and Facebook only marginally elaborated on points already outlined in the joint statement.

Twitter has faced constant criticism over its opaque standards for removing accounts that harass others or spread messages of hate, but experts said a certain level of privacy could be beneficial for the companies as they try to keep up with whatever extremists post.

"When you're in this conflict with extremist organizations online, each side is trying to stay one step ahead of the other," Jasmine El-Gamal, an expert on religious extremism at the Brent Scowcroft Center on International Security, told Mashable. If extremists know the standards Facebook, Google, Twitter and Microsoft have set, she said, then it's easier for them to subvert those standards.

Other parts of the arrangement are somewhat more perplexing, and Twitter again provides a good example why. Twitter has become adept at suspending thousands of ISIS-linked accounts, which is often like playing some warped digital whack-a-mole, as suspended accounts can quickly resurface with a variation of their old name. For example, one devoted ISIS follower, who went by Baqiya Shoutout, returned again and again after suspensions, using "Baqiya" followed by a different series of numbers.

Clearly, these companies view ISIS's brand of Sunni Muslim jihadism as within the type of "terrorist content" they want to combat. But what else do they consider extremist? Twitter has, for example, banned Richard Spencer, a white nationalist who has called for "peaceful ethnic cleansing," but left up the account of Jared Taylor, who founded a think tank to support "white superiority." Without knowing how these companies define extremism or "terrorist," it's hard for the public to say what they're looking for.

"I find it interesting whenever someone says they're looking at 'extremist' accounts, because what's actually getting cracked down?" asked Phillip Smyth, a researcher at the University of Maryland who focuses on armed Shia Muslim groups. "Is it just ISIS? Is it a bunch of white supremacist nuts?"

Smyth said he often looks at a variety of content from armed Shia Muslim militia groups that he considers extreme, though those groups don't have the international pull of ISIS nor the wide range of social media penetration and are therefore not the focus of companies such as Facebook and Twitter. "They're extremists, but they're the kinds of extremists that we just kind of ignore," he said.

It's unclear how much of this information the tech partnership will share with intelligence agencies, if any. If they eliminate videos, posts and images without informing government agencies where those posts originated, there's a chance it could hinder the government's ability to track down those origin points, and hamper the ability of folks such as Smyth to study the online behavior of extremist groups, and to discover who is behind extremist accounts.

"What happens if we're just zapping these accounts before they're up?" Smyth asked. The joint announcement said only that "each company will continue to apply its practice of transparency and review for any government requests."

But it seems unlikely that these companies wouldn't be in touch with government officials about the particulars of the forthcoming database. The White House and others have been in touch with companies such as Facebook and Twitter this year over how technology companies can better combat extremism online. They've found themselves under pressure to eliminate extremist accounts, and Facebook has stepped up efforts to promote anti-radical content as a counter-measure aimed at extremism.

Once Facebook, Twitter, Google and Microsoft have built the database, they plan to figure out how to add other companies to the fold, in the hopes of squashing the ability of extremists and their messages to slink away from one social network to the next. The extremists, meanwhile, will continue searching for an online space not equipped with the tools necessary for substantial censorship.

最新评论