Blindingly amplifying views or posts on social media is one of the key reasons for the rapid spread of misinformation. Over the years, prominent figures have posted or retweeted false information on Twitter. The social network is now giving a chance to withdraw a retweet for such instances through a new Community Notes — its crowdsourced fact-checking program — feature.
Twitter will now notify users if a tweet they liked, retweeted, or replied to receives contextual information from Community Notes contributors. “This helps give people extra context that they might otherwise miss,” the company said in a tweet.
Liking, retweeting, or replying to a tweet gives it some sort of relevance in recommendation algorithms — and Twitter has been pushing its algorithmic feed for a while now. If a Community Notes contributor’s context can debunk the original view, chances are that people might remove their likes or retweets.
The program was first introduced under the name “Birdwatch” in 2021 for US-based users. After Elon Musk bought Twitter, he renamed the program to “Community Notes” — even as Jack Dorsey thought it was the “most boring Facebook name ever.”
In December, Twitter started showing Community Notes to all users across the globe. In January, the company began to accept notes from contributors based in the UK, Ireland, Australia, and New Zealand apart from the US. So the global context is still missing from the Community Notes program.
Over the last few months, Musk & co. have made changes to the algorithm like tweaking the visibility of low-quality notes, extending the type of notes for contributors, and stabilizing the impact score of contributions.
Given that the new management has laid off several people including contractors working in the safety and trust department, Twitter’s reliance on algorithms and crowdsourcing for content moderation has increased rapidly. This burden could further increase as the company has shut off free API access to researchers, a lot of whom contributed to pointing out hate speech and misinformation on the platform.