2021 Facebook leak

January 6 United States Capitol attack
A gallows (top) and a crowd of attackers amid a cloud of tear gas (bottom) outside the Capitol on January 6, 2021
Timeline • Planning
Background
2020 presidential election and other causes 2020–21 presidential election protests 2021 Electoral College vote count (alternate electors) Attempts to overturn the election Democratic backsliding in the US QAnon Republican reactions to Donald Trump's claims of election fraud Social media use by Donald Trump Trumpism
Related groups and persons Ali Alexander Ray Epps Nick Fuentes Rudy Giuliani Jericho March Alex Jones Sedition Caucus Donald Trump Donald Trump Jr.
Participants
Notable people Ashli Babbitt Joe Biggs Jacob Chansley Derrick Evans Tim "Baked Alaska" Gionet Simone Gold Klete Keller Ethan Nordean Rick Saccone Jon Schaffer John Earle Sullivan
Organizations Boogaloo movement Groypers Oath Keepers Proud Boys Three Percenters
Law enforcement response Capitol Police Eugene Goodman Brian Sicknick Howard Liebengood Steven Sund Department of Defense Michael C. Stenger Paul D. Irving Christopher C. Miller Ryan D. McCarthy Walter E. Piatt Charles A. Flynn Daniel Hokanson William J. Walker DC National Guard DC Muriel Bowser Robert Contee Michael Fanone Jeffrey L. Smith Virginia Ralph Northam
Aftermath
Biden inauguration 2021 inauguration week protests Security preparations
Investigations and charges Justice Department investigation Criminal proceedings list January 6 commission House Select Committee public hearings Smith special counsel investigation federal prosecution of Donald Trump
Corporate actions List of companies that halted political contributions Social media suspensions of Donald Trump permanent suspension Suspensions of other social media accounts shutdown of Parler Facebook
Reactions Domestic Antifa culpability conspiracy theory International
Impeachment Second impeachment of Donald Trump trial

The New York Times pointed to internal discussions where employees raised concerns that Facebook was spreading content about the QAnon conspiracy theory more than a year before the 2020 United States elections. After the election, a data scientist mentioned in an internal note that 10 percent of all U.S. views of political content were of posts alleging that the election was fraudulent.[19] Among the ten anonymous whistleblower complaints Whistleblower Aid filed with the SEC on behalf of Haugen, one complaint alleged that Facebook misled the company's investors and the general public about its role in perpetuating misinformation related to the 2020 elections and political extremism that caused the January 6 United States Capitol attack.[1][4] Haugen was employed at Facebook from June 2019 until May 2021, starting within the company's Civic Integrity Team that was focused on investigating and addressing worldwide elections issues on the platform, as well as how the platform could be used to spread political disinformation and misinformation, to incite violence, and be abused by malicious governments until the company dissolved the team in December 2020.[20][21]

In the weeks after the 2020 U.S. presidential election, Facebook began rolling back many content policy enforcement measures it had in place during the election despite internal company tracking data showing a rise in policy-violating content on the platform, while Donald Trump's Facebook account had been whitelisted in the company's XCheck program.[21][22] Another of the whistleblower complaints Haugen filed with the SEC alleged that the company misled investors and the general public about enforcement of its terms of service due to such whitelisting under the XCheck program.[1][4] Haugen was interviewed by videoconference by the U.S. House Select Committee on the January 6 Attack in November 2021 about her tenure at Facebook, the company documents she provided to Congress, the company's corporate structure, and her testimony before Congress the previous month, but none of the information she provided to the Committee was included in its final report.[23][24]

The Files show that Facebook (now Meta) has been conducting internal research of how Instagram affects young users for the past three years. While the findings point to Instagram being harmful to a large portion of young users, teenage girls were among the most harmed. Researchers within the company reported that "we make body issues worse for one in three teenage girls". Furthermore, internal research revealed that teen boys were also affected by negative social comparison, citing 14% of boys in the US in 2019.[25] Instagram was concluded to contribute to problems more specific to its app use, such as social comparison among teens.[26] Facebook published some of its internal research on September 29, 2021, saying these reports mischaracterized the purpose and results of its research.[27]

The Files show that Facebook formed a team to study preteens, set a three year goal to create more products for this demographic, and commissioned strategy papers about the long-term business prospects of attracting the preteen demographic. A 2020 document from Facebook states: "Why do we care about tweens?" and answers that question by saying that "They are a valuable but untapped audience."[28]

An internal memo seen by the Washington Post revealed that Facebook has been aware of hate speech and calls for violence against groups like Muslims and Kashmiris, including posts of photos of piles of dead Kashmiri bodies with glorifying captions on its platform in India. Still, none of their publishers were blocked.[29] Documents reveal Facebook has responded to these incidents by removing posts which violate their policy, but has not made any substantial efforts to prevent repeat offenses.[29] As 90% of monthly Facebook users are now located outside of the US and Canada, Facebook claims language barriers are one obstacle that is preventing widespread reform.

In 2015, in addition to the Like button on posts, Facebook introduced a set of other emotional reaction options: love, haha, yay, wow, sad and angry.[30] The Washington Post reported that for three years, Facebook's algorithms promoted posts that received the 'angry' reaction from its users, based on internal analysis showing that such posts lead to five times more engagement than posts with regular likes. Years later, Facebook's researchers pointed out that posts with 'angry' reactions were much more likely to be toxic, polarizing, fake or low quality.[31]

In 2018, Facebook overhauled its News Feed algorithm, implementing a new algorithm which favored "Meaningful Social Interations" or "MSI". The new algorithm increased the weight of reshared material - a move which aimed to "reverse the decline in comments and encourage more original posting". While the algorithm was successful in its efforts, consequences such as user reports of feed quality decreasing along with increased anger on the site were observed. Leaked documents reveal that employees presented several potential changes to fix some of the highlighted issues with their algorithm. However, documents claim Mark Zuckerberg denied the proposed changes due to his worry that they might cause less users to engage with Facebook. Documents have also pointed to another 2019 study conducted by Facebook where a fake account based in India was created and studied to see what type of content it was presented and interacted with. Results of the study showed that within three weeks, the fake account's newsfeed was being presented pornography and "filled with polarizing and graphic content, hate speech and misinformation", according to an internal company report.[32]

Politico quotes several Facebook staff expressing concerns about the company's willingness and ability to respond to damage caused by the platform. A 2020 post reads: "It's not normal for a large number of people in the 'make the site safe' team to leave saying, 'hey, we're actively making the world worse FYI.' Every time this gets raised it gets shrugged off with 'hey people change jobs all the time' but this is NOT normal."[33]

In 2019, following concerns about Facebook and Instagram being used to trade maids in the Middle East, Apple threatened to remove their iOS apps from the App Store.[34]

The documents have shown a private program known as "XCheck" or "cross-check" that Facebook has employed in order to whitelist posts from users deemed as "high-profile". The system began as a quality control measure but has since grown to protect "millions of VIP users from the company's normal enforcement process". XCheck has led to celebrities and other public figures being exempt from punishment that the average Facebook user would receive from violating policies. In 2019, football player Neymar had posted nude photos of a woman who had accused him of rape which were left up for more than a day. According to The Wall Street Journal, "XCheck grew to include at least 5.8 million users in 2020" according to Facebook's internal documents.[35] The goal of XCheck was "to never publicly tangle with anyone who is influential enough to do you harm".[36]

In 2020, Vietnam's communist government has threatened to shut down Facebook if the social media company does not co-operate on censoring political content in the country, Meta's (then known as Facebook) big market in Southeast Asia.[37] The decision to comply was personally approved by Mark Zuckerberg.[38][39]

In 2021, Facebook developed a new strategy for addressing harmful content on their site, implementing measures which were designed to reduce and suppress the spread of movements that were deemed hateful. According to a senior security official at Facebook, the company "would seek to disrupt on-platform movements only if there was compelling evidence that they were the product of tightly knit circles of users connected to real-world violence or other harm and committed to violating Facebook's rules". As part of their recently coordinated initiative, this included less promotion of the movement's posts within users' News Feed as well as not notifying users of new posts from these pages. Specific groups that have been highlighted as being affected by Facebook's social harm policy include the Patriot Party, previously linked to the Capitol attack, as well as a newer German conspiracy group known as Querdenken, who had been placed under surveillance by German intelligence after protests it organized repeatedly "resulted in violence and injuries to the police".[40]

According to The Wall Street Journal, documents show that in 2019, Facebook reduced the time spent by human reviewers on hate-speech complaints, shifting towards a stronger dependence on their artificial intelligence systems to regulate the matter. However, internal documents from employees claim that their AI has been largely unsuccessful, seeing trouble detecting videos of cars crashing, cockfighting, as well as understanding hate speech in foreign languages.[41] Internal engineers and researchers within Facebook have estimated that their AI has only been able to detect and remove 0.6% of "all content that violated Facebook's policies against violence and incitement".

In December 2021, news broke on The Wall Street Journal pointing to Meta's lobbying efforts to divide US lawmakers and "muddy the waters" in Congress, to hinder regulation following the 2021 whistleblower leaks.[52] Facebook's lobbyist team in Washington suggested to Republican lawmakers that the whistleblower "was trying to help Democrats," while the narrative told to Democratic staffers was that Republicans "were focused on the company's decision to ban expressions of support for Kyle Rittenhouse," The Wall Street Journal reported. According to the article, the company's goal was to "muddy the waters, divide lawmakers along partisan lines and forestall a cross-party alliance" against Facebook (now Meta) in Congress.[53]