What is Meta’s third-party fact-checking programme?
It is an initiative launched by Meta following the surge of disinformation during the 2016 U.S. elections, with the aim of combating hoaxes circulating on its platforms. Since then, the programme has been implemented in more than 100 countries in collaboration with fact-checking teams certified by the International Fact- Checking Network and the European Fact-Checking Standards Network (EFCSN), two international alliances of independent fact-checkers of which Maldita.es is a member. The programme began with Facebook and has since expanded to Instagram, Threads and WhatsApp, all Meta platforms.
Why is Maldita.es part of the third-party fact-checking programme?
At Maldita.es, we work every day to help clean up the information ecosystem from hoaxes and manipulation so that citizens can form opinions based on verified data and facts. Given that in Spain 48% of the population gets its news through social media, according to the Reuters Institute Digital News Report 2024, we believe it is important to monitor and verify potential disinformation circulating on these platforms, a task we have been carrying out for several years.
However, the partnership with Meta allows us not only to independently verify false content: the platforms notify users who have shared a hoax when Maldita.es debunks it and state that they reduce the circulation of content we flag as false. This allows us to have greater impact and reach more people who may be exposed to misinformation.
Do the platforms have any control over Maldita.es’ fact-checks?
Facebook, WhatsApp, Instagram and Threads have absolutely no control over what Maldita.es verifies or how it verifies disinformation on their platforms. Maldita.es will continue to carry out its fact-checks as it always has, following its methodology.
How does Maldita.es find content to fact-check on Meta’s platforms?
At Maldita.es we have access to a Meta platform that includes a list of potentially misleading content circulating on Facebook, Instagram and Threads that may be verified. This list is created based on several factors:
-
User ratings on Facebook, Instagram and Threads: if users consider a post unreliable, they can flag it as “false news”.
-
Content assessments carried out by the artificial intelligence used by Facebook.
-
Journalists at Maldita.es can independently add posts that our community reports as suspicious, even if Facebook’s algorithm has not identified them.
In any case, this list is only a suggestion of what users and algorithms identify as potentially false. Maldita.es can decide to verify only the content it manually adds to the list and not what the platforms’ algorithms flag.
In addition, Maldita.es has a WhatsApp chatbot to report potentially disinforming content. Users save our phone number, activate the chatbot with a simple “hello”, and can send any type of content in any format, from a WhatsApp chain message to a PDF. If we already have an article addressing that content, the chatbot sends it in a semi-automated way; if not, we still log the submission and incorporate it into our workflow so we can respond if we later publish a debunk.
Here you can read more about how to use Maldita.es’ WhatsApp chatbot.
What do you fact-check?
We can fact-check links, images and videos, which may be classified into up to six categories: false, altered photo or video, partially false, missing context, true, satire, not applicable and not rated.
Maldita.es will continue to apply its methodology rigorously to all debunks, as it has always done.
What happens when Maldita.es rates content as false, altered, partially false or missing context?
When Maldita.es or any other independent fact-checker in the programme labels a post as false, altered photo or video, or partially false, Meta states that it takes two actions:
-
It reduces the distribution of the content, meaning it will appear less frequently on Facebook, Instagram and Threads. According to Meta, content is only removed if it violates its Community Standards or misinformation policies.
-
For users, when someone attempts to share content considered false, they are alerted and shown the debunk or explanatory article written by Maldita.es. In addition, users who have already shared the hoax receive a notification informing them that the content is false. It is then up to the user to decide whether to delete it.
In the case of WhatsApp, we can send the verification directly to users, but as it is an encrypted platform, Meta states that it does not influence user messages because it cannot see them.
What happens when an account or profile repeatedly publishes disinformation identified as false?
Meta will reduce the distribution of pages, groups, profiles, websites and Instagram accounts that publish hoaxes and disinformation. Those that do so repeatedly may lose the ability to monetise, publish ads or register as a news page on Meta. Independent fact-checkers like Maldita.es do not know how many negative ratings are required for these measures to be applied, nor how long it takes for them to be lifted if a site stops spreading false information, as Meta has not specified this.
Can owners of accounts or profiles whose content is marked as “false”, “altered” or “missing context” appeal?
Yes. Accounts that have shared content marked as “false” are automatically notified by the platform and can review the fact-check and respond to Maldita.es’ assessment. This response must be submitted within one week of notification and can be done in two ways:
-
By sending an email to [email protected] explaining why the rating is incorrect and always providing a link to the content. Maldita.es can only review content that has been fact-checked by our team.
-
Through the platforms themselves. On Instagram and Threads, this can be done via the “Request review” option in the fact-check notification at the bottom of the page. On Instagram, appeals can also be made through Account Status, both on desktop and in the app, as Instagram explains.
On Facebook, a review can be requested from a profile via “Profile status”, from a page the user owns via “Page status”, from an administered group via “Group status”, and for ads via the “Business Help Center”, as Facebook explains.
Appeals can take two forms:
-
A review, if the user has corrected the content or ad marked as false. To ensure fair and consistent handling of corrections, we have developed a correction policy, which you can read here.
-
A dispute, if the user believes the rating is incorrect and clearly explains why.
If a correction or dispute is successful, the content penalty will be removed and measures against the page or domain will be lifted. You can read more about this process here.
Does Maldita.es have access to Meta users’ personal data?
No. Under no circumstances does Maldita.es have access to Facebook users’ personal data.
Does Meta pay Maldita.es for this service?
Like all fact-checkers participating in this programme, Maldita.es is paid for the service it provides to the platform. You can find more details in our Malditas Cuentas.
Do you at Maldita.es believe this helps reduce disinformation?
Any effort by the various actors involved in this issue helps move towards a possible solution. While the presence of independent fact-checkers on Meta may help reduce hoaxes or at least help citizens identify them more quickly, it is necessary for Meta and other platforms to share data transparently on how disinformation circulates and on the impact of initiatives such as third-party fact-checking. This transparency is essential to realistically assess their impact and determine next steps in the fight against falsehoods.
At Maldita.es we also believe it is important not only to identify and debunk hoaxes so citizens are better informed, but also to equip people with tools that allow them to identify, verify and debunk disinformation themselves. We invite you to take our challenge to detect hoaxes through our WhatsApp chatbot.
If you still have any questions about this agreement with Facebook or any other Maldita-related issue, you can contact us at [email protected].
This page was updated on 17/10/2024.