Twitter for years functioned as an unrestricted mouthpiece for hackers of all stripes, from freewheeling hacktivists like Anonymous to the Kremlin-created cutouts like Guccifer 2.0. But as the company tries to crack down on hackers' use of its platform to distribute their stolen information, it's finding that that's not a simple decision. And now, less than three weeks before Election Day, Twitter has put itself in an impossible position: flip-flopping on its policy while trying to navigate between those who condemn it for enabling data thieves and foreign spies, and those who condemn it for heavy-handed censorship.
On Thursday evening, Twitter's head of trust and safety, Vijaya Gadde, posted a thread of tweets explaining a new policy on hacked materials, in response to the firestorm of criticism it received—largely from the political right and President Donald Trump—for its decision to block the sharing of a New York Post story based on alleged private data and communications of presidential candidate Joe Biden's son, Hunter Biden. Gadde wrote that the company was taking a step back on its "Hacked Materials Policy." The company will now no longer remove tweets that contain or link to hacked content "unless it is directly shared by hackers or those acting in concert with them," Gadde wrote. Instead, the company will "label Tweets to provide context."
Despite that new rule, links to the Post article initially remained blocked, because it also violated Twitter's policy on sharing private personal information, another spokesperson for Twitter posted last night. But Twitter ultimately backed down from that stance too, allowing the story to circulate as it broadly rethought its treatment of posts about hacked information.1 "Why the changes?" Gadde wrote. "We want to address the concerns that there could be many unintended consequences to journalists, whistleblowers, and others in ways that are contrary to Twitter’s purpose of serving the public conversation."
Rather than solve Twitter's hacked data dilemma, though, Twitter's backpedaling on its policy has only highlighted just how stuck it is between impossible options, says Clint Watts, a disinformation-focused senior fellow at the Center for Cyber and Homeland Security at George Washington University and author of the book Messing With the Enemy. And it may also leave Twitter open to exploitation by a well-crafted hack-and-leak operation, just as Russian hackers carried out in 2016.
"It's a super difficult problem to thread," Watts says. "If they didn’t take that down, and it turns out to be a foreign op, and it changes the course of the election, they’re going to be right back testifying in front of Congress, hammered with regulation and fines." After all, Twitter faced widespread criticism for allowing itself to be exploited ahead of the 2016 election by Kremlin hackers who distributed information stolen from the Democratic National Committee and the Clinton campaign, as well as by disinformation trolls working for the Kremlin-backed Internet Research Agency.
In response to those incidents, Twitter implemented its rule against the “distribution of hacked materials” in 2018, which banned posting hacked content directly or linking to other sites that hosted it. Critics of the policy, however, argued that it also risked blocking legitimate news stories in the public interest if they are based on information released without authorization.
"There's incredible journalism that starts with hacked materials," says Lorax B. Horne, editor in chief of the whistle-blowing "leaks" group known as Distributed Denial of Secrets, or DDoSecrets.2 DDoSecrets published a massive collection of internal memos, financial records, and other data stolen from 200-plus police organizations in June, and told WIRED that the information had been given to them by a transparency-focused hacker affiliated with Anonymous. Journalists dug through the material and found revealing stories about police misperceptions of antifa and Homeland Security surveillance practices, including those focused on Black Lives Matter protestors.
Twitter responded by banning DDoSecrets' account, suspending the group's members who promoted the links to the hacked police documents, and blocking anyone from tweeting links to it. Despite the comments on Thursday from Twitter's Gadde about trying to avoid "unintended consequences to journalists, whistleblowers and others," the ban remains in place today.
Horne points out that Twitter's new policy of removing tweets or banning Twitter users who are "acting in concert" with hackers is vague. Does it apply, for instance, to the International Committee of Investigative Journalists for its publication of the Panama Papers, a groundbreaking collection of leaked documents that revealed widespread money laundering? The Panama Papers relied on an anonymous source of data taken from a Panama law firm that claimed to have been hacked. "Journalists who are self-aware and honest have to recognize that hackers are often doing work that we need," says Horne.
Twitter's rulemaking is muddied further by the fact that the case that led to its most recent changes is itself still far from clear: The New York Post has claimed that its collection of Hunter Biden's communications and photos came from a laptop he abandoned at a repair shop, not from a hacking operation. But The New York Times reported Wednesday that US intelligence analysts have been concerned about the possibility of a Russian hack-and-leak operation focused on Hunter Biden and his ties to the Ukrainian oil firm Burisma. It’s unclear why Twitter initially invoked the hacked materials policy for banning the Post story. Its decision may have been based on nonpublic evidence, or perhaps a mistaken assumption based on the fear of Kremlin involvement. Twitter didn't respond to WIRED's request for comment.
Facebook, for its part, has also banned BlueLeaks and restricted the sharing of the Post's Biden story. Unlike Twitter, however, Facebook hasn't backed down from those decisions—leading to plenty of cries of censorship, but fewer accusations of inconsistency in its hacked-info rules.
Twitter's apparent policy reversal comes after intense political pressure from right-wing voices, including a tweet from President Trump. Watts points to that reversal as an example of "gaming the ref," a fear he's heard broadly from social media companies: that they'll be bullied by politicians accusing them of bias. "The system can’t figure out the game, because the regulators—the politicians—are part of the problem," he says.
"I'm just glad I'm not the head of trust and safety at Twitter," Watts says. "Because I don't know what policy I would craft that would make everyone happy, keep me out of the regulatory crosshairs, and still allow for the free flow of information."
1Updated 10/16/20, 5:40pm ET to note that Twitter had now fully lifted its ban on the New York Post story.
2Updated 10/16/2020, 3:30 pm ET: This story has been updated to correct Lorax B. Horne's position at DDoSecrets. Horne is the editor in chief, not a cofounder as originally stated.