Twitter launches Birdwatch, a fact-checking program that aims to combat misinformation

Twitter launched its Birdwatch program, which aims to resolve misinformation on the platform by allowing users to check tweets, the company announced Monday. Users of the pilot program, which will initially include around 1,000 users in the U.S., will be able to add notes to tweets to provide context.

For now, users participating in the pilot can write notes in individual tweets, but the notes will not be publicly visible on Twitter itself, only on the public site Birdwatch. Pilot users can also rate notes submitted by other program participants.

Here is an example of what a tweet with Birdwatch notes would look like:

Image: Twitter

“We believe that this approach has the potential to respond quickly when misleading information spreads, adding a context that people trust and find valuable,” wrote Twitter product vice president Keith Coleman in a blog. “Eventually, our goal is to make annotations directly visible in Tweets to the global Twitter audience, when there is consensus from a wide and diverse set of contributors.”

Twitter first confirmed that it was working on Birdwatch last year, but it was not expected to launch before the U.S. presidential election.

Like most social media companies, Twitter fought the spread of misinformation and advertising on its platform. The company took steps to try to combat electoral misinformation during the U.S. presidential campaign and then to label the tweets with wrong or misleading information about the election. The program had mixed results, however, and did not appear to act as a deterrent for former President Trump, whose account was permanently suspended in January.

Twitter says it interviewed more than 100 people across the political spectrum, who told the company that Birdwatch’s notes provided a useful context for better understanding tweets. “Our goal is to build Birdwatch openly and have it shaped by the Twitter community,” wrote Coleman.

All data contributed to Birdwatch will be available and downloadable in TSV files, and Twitter will publish the algorithms that feed the program publicly in a Birdwatch Guide, the company said. The initial rating system is available on Twitter’s GitHub page.

“We know that there are a number of challenges to building a community-driven system like this – from making it resistant to attempts at manipulation to ensuring that it is not dominated by a simple or biased majority based on the distribution of employees,” noted Coleman. “We will be focused on these things throughout the pilot.”

To sign up to participate in Birdwatch, users can sign up here.