How Meta’s Fact-Checking Programme Works

Paris:

Internet giant Meta’s fact-checking programme aims to counter false information on its social platforms — Facebook, Instagram and Threads — with the help of media organisations, including AFP.

Here is a guide to how it works, following Meta chief Mark Zuckerberg’s announcement on January 6 that he was halting the programme in the United States.

Who does fact-checking?

Meta has since 2016 tasked journalists with assessing false information on its platforms under its “third-party fact-checking” programme — which currently has 80 media organisations working on it.

In the United States, for example, Meta has 10 media partners for the programme, including AFP, USA Today and specialist fact-check sites Lead Stories and PolitiFact.

The partners are certified by the International Fact-Checking Network (IFCN), set up in 2015 by the Poynter Institute, a US-based non-profit media training and advocacy group.

To be certified by the IFCN, media organisations must meet its standards of editorial quality, neutrality and independence.

Zuckerberg said he will replace the fact-checks with user-generated “community notes” similar to those used on rival platform X, formerly Twitter, owned by Elon Musk.

What do fact-checkers do?

AFP publishes fact-check articles in 26 languages, aimed at readers of posts published in those tongues.

The articles explain why claims are misleading and provide correct information for readers.

To start their articles, AFP’s 150 fact-checkers identify claims that are misleading, potentially dangerous and circulating widely on the platforms.

They verify facts, not opinions or beliefs.

They do so by assembling concrete evidence that is transparently sourced and cross-checked.

They explain the steps of their investigation and publish weblinks to their sources where possible.

The media organisation that produces a fact-check article publishes it on their own website.

In the case of AFP fact checks published in English, this site is factcheck.afp.com.

How does Meta use fact-checks?

To apply a fact-check to a claim circulating online, the media organisation enters the web address of its article into an interface provided by Meta.

This interface associates the article with the social media post containing the false or misleading information.

This causes the misleading post to become less widely viewed on the social platform and a rating such as “false” or “misleading” to appear alongside the post.

The post is not removed from the platform — instead, a link to the fact-checking article appears under the post, directing readers to the verified information.

Users who have shared the misleading post receive a notification providing them with a link to the fact-check article.

If the author of the post corrects it, the rating is removed and the restriction on how many users see the post is lifted.

When are fact-check articles paid for?

Meta pays the media organisations for publishing the articles that are linked to the misleading posts.

The media organisations may write articles on any subject they choose — even concerning misinformation that is not circulating on Meta’s platforms.

But Meta only pays them for the articles that are linked to specific posts on its platforms.

Meta does not apply ratings or attach fact-checks to posts by political figures.

This does not prevent the media organisations from publishing articles about such posts — as AFP regularly does. But they do not receive payment from Meta for those articles.

Fact-checkers are not to be confused with Meta’s content moderators.

Moderators remove posts that break a platform’s rules, for example ones that feature pornography or racist content, or incite terrorism or violence.
 

(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)