Twitter has turned to a new resource to help combat the spread of disinformation on the platform: its own users. The social media giant has revealed Birdwatch, a crowdsourced fact-checking scheme that will have verified users flagging tweets for false or misleading content.
Birdwatch, which has been undergoing private trials for some time, also allows these users — acting in a moderator role — to discuss flagged tweets in a separate section of the site, and add notes to tweets detailing their factual issues. The first full pilot scheme is launching imminently.
The project is Twitter’s latest move in an ongoing attempt to prevent potentially harmful disinformation and misinformation from spreading on its turf, a problem that worsened significantly in 2020 with the COVID-19 pandemic and the U.S. presidential election.
Twitter has previously taken the unilateral steps of applying content warnings to misleading tweets, including those of former president Donald Trump, and banning tens of thousands of accounts believed to be involved in organized disinformation campaigns. Birdwatch represents the first time Twitter has involved other users in its efforts.
The initial number of Birdwatch users will be small, though anyone with a verifiable email and phone number with a trusted U.S. carrier may apply to join the pilot. Notes on flagged tweets will also only be visible to other Birdwatch users for the time being — there are none currently visible on this public example — but in time Twitter hopes to make these notices visible to everyone, with Birdwatch discussion threads also readable as a form of public fact-checking service.
Coming from the mayor’s office soon. They want to convert all of our water fountains from still to sparkling! pic.twitter.com/c4Fwb75lLBJanuary 23, 2021
This visible, community-driven approach was apparently effective at building trust in the system, according to the Twitter blog. Although Birdwatch is still a way from rolling out in full, it’s clearly intended to be a more transparent alternative to the company’s own, top-down fact-checking, similar to how sites like Wikipedia and Reddit are moderated.
This will, in turn, require a degree of self-moderation to prevent Birdwatch users mislabelling truthful tweets as misleading. This explains why Twitter makes users apply to the program and makes Birdwatch discussions publicly readable — for vetting and transparency.
Twitter, for its part, seems aware of potential shortcomings but is willing to try such a new approach. “We know this might be messy and have problems at times,” wrote vice president of product Keith Coleman in the blog post, “but we believe this is a model worth trying.”