Meta’s decision to pull the plug on its independent fact-checking system in the US received sharp criticism from various quarters, and sparked a wide-ranging debate about how the social media giant is dealing with misinformation on social media.
In a blog post titled ‘More Talk, Fewer Errors’ published on January 7, the company behind Facebook, Instagram, WhatsApp, and Threads announced a series of changes to the way content will be handled across its apps in the US.
A tangible change to Meta’s content moderation approach is that it will remove fact-checking of posts by fact-checkers in the US, replacing them with a ‘Community Notes’ system similar to X (formerly Twitter).
Why is Meta moving from fact-checking to public notes? How will they work? What are the potential drawbacks of public notes? Have they been successful in the past? Take a look.
What caused the change
Nine years ago, Meta began flagging fake news with the help of external auditors. Its independent fact-checking program has been expanded following reports that Russian disinformation campaigns targeted American voters to influence the 2016 US presidential election.
To this end, fact-checkers and experts accredited by the International Fact-Checking Network (IFCN) can independently review and evaluate possible misinformation in Meta forums by citing their original report, interviewing primary sources, consulting public data, and conducting media analysis, including pictures. and videos.
Meta has ensured that all content rated as false by IFCN verified fact checkers will be less visible to users. It may also affix a warning label below such content, linking to an article published by the fact-checker.
Now, Meta said this approach is flawed. “Experts, like everyone else, have their biases and opinions. This was reflected in the decisions others made about what to test and how,” the company said.
However, the IFCN argued that the fact-checkers were not biased in their work. “The fact-checkers used by Meta follow a Code of Ethics that requires impartiality and transparency,” IFCN director Angie Holan said in a statement.
Besides criticizing Meta for throwing fact-checkers under the bus, Holan said their decision “came after a lot of political pressure from the new administration and its supporters.”
US President Donald Trump has criticized major technology companies for allegedly targeting conservatives online.
How Public Notes work
Community Notes was first piloted as a program called ‘Birdwatch’ by Twitter in 2021 that is before Elon Musk bought the platform for $46 billion and renamed it X.
The crowdsourced fact-checking model allows users to add facts and context under specific posts. A Community Note appears below a post only if enough contributors vote that the content it provides is useful.
As a result, the model is said to improve as more users participate.
Currently, anyone on X can be a contributor and add Public Notes as long as they meet certain criteria such as having an account that is six months old, a verified phone number, and zero violations of X’s rules.
Initially, contributors are only allowed to rate Public Notes as useful or not. Over time, they are allowed to write and paste their own Social Notes to be rated by other participants.
All Public Notes contributions to X are publicly available. Anyone can download data to analyze trends and flag problems.
Meta’s Community Notes model will likely be similar to X’s. “It will require consensus among people with multiple viewpoints to help prevent biased ratings,” the company said.
US users have been able to sign up to contribute via Facebook, Instagram, and Threads since January 7.
Challenges with public notes
Given their multi-source nature, the Public Notes may be vulnerable to aggregate fraud. To address this challenge, X uses a clustering algorithm to determine whether a note appears at the bottom of a post.
This means that a Note will only be shown if it is rated as ‘helpful’ by people who tend to disagree in their previous ratings, according to X’s guidelines.
UX says this method of integration “helps prevent one-sided ratings and prevent one group from being able to vote en masse to decide which notes are displayed.”
To ensure diversity of opinion, X said he quickly asks contributors who “may have a different opinion” of their entry through the ‘Need Your Help’ tab within the Note.
Donors of public notes are also given automatically generated nicknames so that they are not harassed with their donations.
Contributors who write too many Community Notes rated ‘Not Helpful’ will be temporarily banned to prevent the system from being flooded with spam, low quality Notes.
However, some challenges remain. Community Notes may not be effective in stopping the spread of misinformation on other platforms. Their ability to capture the nuance that goes into fact-checking is also questioned.
Discover the Benefits of Our Subscription!
Stay informed with access to award-winning journalism.
Avoid misinformation with reliable, accurate reporting.
Make smart decisions with important details.
Choose your subscription package