Social Media Election Policy Tracker

Axios reports that Meta will begin to remove expired political ads from the Ad Library rather than keeping them in the archive: "Meta, then Facebook, introduced its Ad Library on May 24, 2018. At the time, it said archived ads and advertisers would only be stored in the library for seven years. A spokesperson confirmed that ads will begin to disappear this Saturday."

Meta is terminating its third-party fact-checking program in favor of a new Community Notes program, which allows users to contribute context to posts. This shift also includes reducing restrictions on topics such as immigration and gender identity and modifying enforcement practices to rely less on automated systems and more on user reports for less severe violations. Additionally, Meta plans to reintroduce and personalize the display of political content on its platforms rather than limit it.

According to an Axios article in early November, the ban "was initially supposed to expire at 11:59 p.m. PT on Election Day, but Meta is extending the ban [for a few more days] to prevent any confusion or misinformation from spreading while votes are still likely being counted."

Meta then rolled back that extension until only November 7th. 

Musk-funded PAC created the "Election Integrity Community" on X, a crowdsourced place to "share potential incidents of voter fraud or irregularities you see while voting in the 2024 election." News reports claim widespread false claims proliferated on the site, dying down after Trump decidedly won the election. 

The New York Times, citing Meta employees, reports substantial shifts in the company's election policies. These include CEO Mark Zuckerberg no longer personally meeting with election security leaders, reducing the workforce dedicated to election security, and dissolving the election integrity team. Additionally, Zuckerberg "has also decided not to have a 'war room,' which Meta previously used to prepare for elections. Instead, Meta said it will be running an election operations center closer to the November vote."

Meta announced a ban on RT, Rossiya Segodnya, and other Russian state media networks from its platforms, citing their involvement in covert influence operations using deceptive tactics, escalating Meta's response to Russian State media following prior measures that included blocking ads and limiting the visibility of their posts.

On Instagram, Washington Post reports find Meta reducing political content, including that related to elections and voting.

In a statement in September 2024, Meta said that due to feedback "we have spent the last few years refining our approach on Facebook to reduce the amount of political content – including from politicians’ accounts – you see in Feed, Reels, Watch, Groups You Should Join, and Pages You May Like."

Facebook announces, "We have begun rolling out a Political Content setting on Facebook via Manage Defaults. Changing from the default setting may increase the amount of political content people see from accounts they follow, as well as increase political content recommendations from accounts they don’t already follow." 

Grock, Twitter's AI election chatbot, spread misinformation by saying that new candidates could not be added to the ballots after Biden announced withdrawing from the election.

In letter to congress, Zuckerberg says Meta "no longer temporarily demotes [stories] in the U.S. while waiting for fact-checkers."

In quarterly Adversarial Threat Report Meta says that Russia is a top threat in covert operations regarding election influence in the United States. Meta found six covert influence operations in Q2 originating in Russia and reports shutting them down before they gained traction.  

CrowdTangle, a public insights tool for exploring social media content, was shut down permanently. In its place, Meta offers the Meta Content Library and Content Library API. 

In November 2023 Meta announced a reprise of their political ad block policy a week before elections. In August 2024 Meta announces the specific dates and times of ads blocked: "we will block new social issue, electoral, and political ads during the final week leading up to Election Day in the US (12:01 AM PT on October 29, 2024 through 11:59 PM PT on November 5, 2024). While new ads will not be able to run during this restriction period, ads that have run prior to the restriction period and served at least one impression will be allowed to continue running during this time with limited editing capabilities." 

Meta acknowledged two issues related to the handling of political content on the platform. A fact-check label was incorrectly applied to a real image of former President Trump following an attempted assassination due to similarities with a doctored version. Additionally, Meta AI initially failed to respond correctly to questions about the event, a limitation in how AI chatbots handle rapidly developing news. Meta has since updated the responses and corrected the fact-check label error.

Meta announced the removal of restrictions on former President Donald Trump’s Facebook and Instagram accounts, allowing him to use the platforms without the threat of suspension for violating terms of service, including content that could delegitimize the upcoming election. These restrictions were initially put in place after the January 6, 2021, Capitol attack, where Trump was indefinitely suspended for posts that risked inciting violence. Meta reinstated his accounts in 2023 with some limitations, which are now no longer applicable. 

Elon Musk, owner of X and its most-followed user, endorsed Donald Trump for president in a post. He later hosted a livestreamed conversation with Trump on the platform.

Historically, major platforms and their owners have not typically endorsed political candidates.

The U.S. Department of Justice revealed that nearly 1,000 AI-enhanced Russian bots, linked to Russia’s state-controlled news network, Russia Today, and backed by Russia's Federal Security Service, impersonated Americans and posted misleading content to manipulate public opinion. X suspended the accounts. 

The Oversight Board emphasizes the critical role of social media in shaping elections, with 80 countries holding elections in 2024. While platforms can facilitate transparent elections, they are also tools for spreading disinformation and inciting violence. The Board offers nine key lessons for improving election integrity, such as enhancing enforcement, addressing political speech that incites violence, setting standards for AI-generated content, and increasing transparency. The recommendations call on companies like Meta to protect civic spaces and political expression while upholding human rights.

Meta expanded its policy on manipulated media across Facebook, Instagram, and Threads. Initially, the policy targeted only AI-generated videos that falsely portrayed people saying things they didn’t. The update now covers AI-generated content like audio and photos, based on recommendations from Meta’s Oversight Board. The platform will label manipulated AI content with context instead of removing it, unless it violates other policies such as voter interference or harassment. Labels such as "Made with AI" or more prominent labels will help users assess the content. Independent fact-checkers will continue reviewing AI-generated content marked as false or altered.

In September 2024 the company announced the AI info labels only appear on AI generated content rather than AI modified content. 

When Threads was launched, fact-checkers could not directly fact-check on the platform. This was later updated. As of April 2, 2024: "Our fact-checking program is now live on Threads. Fact-checking partners now have the ability to review and rate misinformation on Threads directly."

Users of Threads be able to opt-in to posts being discoverable on the Fediverse, a collection of decentralized social media platforms that use open standards and protocols. 

Meta announced that CrowdTangle, a data transparency tool for external researchers, will no longer be available after August 14, 2024. In their announcement, Meta stated that new research tools, Meta Content Library and Content Library API, will be available instead. 

In written disclosures to the Senate Judiciary Committee, X reported that it had 3,317 trust and safety employees and contractors in May 2022. By May 2023, this number had decreased to 2,849, reflecting a reduction in staffing over the course of a year.

Meta adjusted its recommendation system to no longer suggest political content on Instagram and Threads. Users can still see political content from accounts they actively follow. Critics raised concerns about the criteria used to define "political" content, arguing that this change limits a vital avenue for voter education. According to Meta's blog post, political content is defined as content "potentially related to things like laws, elections, or social topics."

Meta announced that Facebook, Instagram, and Threads will use common technical standards to identify and label AI-generated photos and videos. This initiative complements a previous decision to label content generated with Meta's AI tools. The labels will include visible markers on the images, as well as invisible watermarks and metadata embedded within the image files. Meta acknowledges that it is currently not possible to identify all AI-generated content, and there are methods to remove invisible markers.

Facebook tells CNBC in September 2024 that they started hosting virtual training sessions for state and local officials prior to the 2024 presidential election. These training sessions cover anything from creating ads, to fostering a safe community online, to using Facebook to reach a broader community. Some state and local officials have said they didn't know about this resource and have a hard time getting in direct touch with Meta. 

Announced in November 2023, but going into effect in January 2024, Meta updated its policy to require AI disclosures in Political Ads. Further, from The New York Times, "Meta added that it would bar advertisers from using its own A.I.-assisted software to create political or social issue ads, as well as ads related to housing, employment, credit, health, pharmaceuticals or financial services."

Meta's Threads follow Instagram's community guidelines but not Meta's regular fact-checking programs. In an update on December 12, Meta explained, "Currently, when a fact-checker rates a piece of content as false on Facebook or Instagram, we extend that fact-check rating to near-identical content on Threads, but fact-checkers cannot rate Threads content on its own. "

In April 2024, Meta launched a fact checking program on Threads.

Meta updates its community standards for voter and census interference: 1) The misrepresentation category now includes "Voter/Census Interference;" 2) The Coordinating harm category continues to include "Voter/Census Fraud."

They also include new policies under "Additional Info/Context Needed to Enforce:" 1) Coordinated Interference is content calling for interference in census or election participation; 2) Law Enforcement Consequences are claims that census or voting participation will lead to arrest, deportation, or imprisonment; and 3) Intimidation at Voting Sites are defined as statements advocating monitoring voters or officials with the intent to intimidate, using militaristic language or expressing control.

Meta released its public-facing plan for the 2024 elections. Meta emphasized continuity with its approach in previous elections of blocking new political ads in the week before the election, though ads previously running can continue. 

Advertisers "will have to disclose when they use AI or other digital techniques to create or alter political or social issues" in advertisements. 

Meta will label state-controlled media. 

Meta introduced new controls that allow users to opt out of its fact-checking program. This change permits debunked posts to appear in participants' news feeds with only a warning label, rather than being suppressed or removed.

After previously promising to expand the Election Integrity team, X, under Elon Musk's leadership, cuts half of the team, including its leaders.

X revised its Civic Integrity Policy to include measures for labeling or removing false or misleading information intended to undermine public confidence in an election or other civic processes. Tweets that violate these policies will have "reduced visibility" (de-amplified) and will be labeled, but not removed entirely.

X announces a new policy allowing political advertising, emphasizing freedom of expression and forthcoming updates to its Civil Integrity Policy. A global advertising transparency center will be created to review political posts. U.S. policies prohibit false or misleading election content, and limit ads to U.S. persons/entities with pre-approval. 

Collaborative research evaluating the impact of Facebook and Instagram on political attitudes and behaviors during the 2020 U.S. election cycle were published in scientific journals Science and Nature. The findings show minimal evidence that the platforms' key features cause harmful polarization or significantly affect political outcomes. However, some external researchers noted that the short data collection period and its timing near the election may limit the findings' broader relevance. More studies are expected to be released in the future. 

Meta has launched Threads. Threads is positioned as a competitor to other text-based social media platforms, offering features that cater to public discussions and private conversations.

Musk rebrands Twitter by removing the bird logo and changing the official name to X.com

A letter from the Senate Committee on Rules and Administration states that Meta has reduced its election-related staff from 300 to 60 employees, representing an 80% decrease. Further layoffs are also planned.

Yoel Roth, Twitter's former Head of Trust and Safety, stated that the platform's Trust and Safety program has become "nonexistent" under Elon Musk's leadership. He expressed concerns about the significant reduction in staff and resources dedicated to maintaining safety and enforcing policies on the platform. Roth's comments reflect broader worries about the ability of Twitter to manage harmful content and maintain user trust amidst substantial organizational changes.

Ella Irwin resigned from her role as Twitter's Head of Trust and Safety after holding the position for seven months, following the resignation of her predecessor.

Meta has made substantial staff reductions, cutting nearly a quarter of its workforce, which initially numbered around 87,000 in 2022. The company announced 11,000 layoffs in November 2022 and a further 10,000 in March 2023. 

X has expanded its Community Notes feature, a crowdsourced moderation tool, to include images. This update comes as concerns about the spread of deepfake images rise. The new feature aims to help users identify misleading or manipulated images by allowing the community to add context and annotations.

Twitter's workforce was reduced by 80% following Elon Musk's acquisition. In 2022, the company employed 7,800 people. Since then, 6,300 employees have been laid off.

Twitter removed state media labels from all accounts, including outlets such as BBC, NPR, and other publicly funded media. This decision followed backlash against the labeling, now allowing content from state-affiliated media, including those from China and Russia, to circulate without designation. 

Twitter adds a message that says "Visibility limited: this Tweet may violate Twitter's rules against Hateful Conduct" as a transparency measure alerting users to the platform's visibility filtering measures against hate speech.

In a letter to Meta employees, CEO Mark Zuckerberg outlined a plan to restructure the company's workforce by increasing the ratio of engineers to other roles. This shift is part of Meta’s "Year of Efficiency" initiative to streamline operations.

Meta announced that it readmitted former President Donald Trump to its platforms after a two-year suspension. The company will monitor his account with "guardrails." The decision followed a review of his ban, which was initially imposed due to the risk of further incitement of violence following the January 6 Capitol riot.

Meta ended contracts for about 200 content moderators in early January 2023 and laid off at least 16 members of Instagram’s well-being group. Additionally, the company cut more than 100 positions related to trust, integrity, and responsibility. These staffing reductions also halted the development of a new fact-checking tool designed to allow third-party fact-checkers to add comments to questionable articles on Facebook.

Following its decision to allow political ads, Twitter announced two new processes: a certification process for ad sponsors and a disclosure process for political ads. The specifics of the certification process are not publicly available. Users can request a disclosure report online to access details about political ads running on the platform, placing the burden on the public to seek this information rather than Twitter providing proactive disclosure.

Twitter has lifted its ban on political ads, with policies varying by country. In the United States, the policy is permissive but prohibits false or misleading content related to elections. This includes misinformation about election participation, content intended to intimidate or dissuade voters, and information that undermines public confidence in elections. The review process for determining what constitutes "false or misleading" was not disclosed.

Twitter suspended, without notice, the accounts of multiple journalists, including CNN’s Donie O’Sullivan, The New York Times’ Ryan Mac, The Washington Post’s Drew Harwell, and independent journalist Aaron Rupar. Twitter claimed the suspensions were due to the journalists' engagement in doxing behavior against Elon Musk. 

Twitter re-emables Twitter Blue verifications, a paid service that allowed users to obtain a blue checkmark, a symbol traditionally reserved for notable public figures, corporations, and influential accounts. 

Twitter’s new Head of Trust and Safety said that under Elon Musk's leadership, the platform is increasingly relying on automated systems to moderate content. This shift includes reducing manual content reviews and focusing on restricting the distribution of certain content rather than removing it entirely.

Meta has laid off the leadership and staff of the Meta Journalism Project, which has not posted any new milestones since March 2022. This move is part of a broader shift in Facebook's approach to journalism, as the company has recently terminated several news-related programs, including Instant Articles and Bulletin. Additionally, Facebook has stopped paying U.S. publishers for their stories, renamed the "News Feed" to "Feed," and announced that its "Top Stories" section will soon be curated by algorithms instead of humans.

Meta updated policies on Facebook and Instagram to allow political ads that challenge results of past elections including the 2020 elections, which can now be described as "stolen" or "rigged." Media outlets reported this policy change nearly a year after it was implemented, noting the quiet shift made after the 2022 midterms without public announcement.

Twitter's new enforcement philosophy focuses on de-amplifying content that violates its terms rather than removing it. The approach is described as “freedom of speech, but not freedom of reach.”

Elon Musk announced a "general amnesty" for suspended Twitter accounts following a user poll on the topic. The amnesty led to the reinstatement of users previously banned for offenses including violent threats, harassment, and spreading misinformation.

Twitter reinstated former President Donald Trump’s account. His account was originally taken down after January 6, 2021.

Musk announced his plans to reverse the ban in May 2022, then announced his reinstatement in a tweet. 

Yoel Roth, Twitter's Head of Trust and Safety, resigned in November 2022. In an op-ed published a week later, Roth emphasized the challenges of maintaining principled content moderation under Elon Musk's leadership, stating that a platform "whose policies are defined by edict has little need for a trust and safety function dedicated to its principled development."

Near the 2022 election, Twitter cut 15% of its Trust and Safety team, including Neema Singh Guliani, who led democracy, national security, and civil rights policy. Other teams critical to content moderation, such as the Product Trust team—which develops policies to prevent new features from harming users—and the Health team—which implements Trust and Safety policies—faced even heavier cuts.

Elon Musk announced that Twitter would form a content moderation council with diverse viewpoints but did not provide timeline or specifics for delivery.

Twitter fired several top officials, including Vijaya Gadde, the head of legal, policy, and safety. Gadde was responsible for the decision to suspend President Trump's account. 

Elon Musk completed his acquisition of Twitter on October 27, 2022, in a deal valued at approximately $44 billion. Following the acquisition, Musk immediately took control of the company, dissolving the board of directors and naming himself CEO. He posted "the bird is freed" to announce the closing of the deal.  

Meta’s Q3 2022 Adversarial Threat Report details the removal of three networks engaged in coordinated inauthentic behavior (CIB) across multiple platforms. Related to U.S. elections Meta updated their findings on December 15, 2022: "Our investigations linked the Russian network to two companies in this country: Structura National Technologies, an information technology firm and Social Design Agency (Агентство Социального Проектирования), a marketing and political consulting firm."

"Meta is merging its business integrity unit, the team that moderates ad content, with its central integrity team, which moderates users' posts, according to an internal memo obtained by Axios." 

Meta disbanded its Responsible Innovation Team, which comprised approximately two dozen engineers, ethicists, and other specialists. The team collaborated with internal product teams, outside privacy experts, academics, and users to identify and address potential concerns and downsides related to new products and changes to Facebook and Instagram.

Facebook explored banning political advertising starting in summer 2021 but abandoned the idea by summer 2022. The decision came amid concerns about the impact of Elon Musk's approach to free speech on Twitter.

Twitter announced the rollout of candidate account labels for the 2022 U.S. elections to prevent the spread of fake accounts. The labels are visible on candidate profiles and in tweets. They also announced a redesign of its misleading information labels for the 2022 elections and "pre-bunks." The pre-bunk prompts appear directly on users' timelines and in search results when people type terms, phrases, or hashtags associated with misinformation. 

Twitter introduced the Election Explore Tab for the 2022 elections, a centralized hub for real-time political conversations, resources, and breaking news. The tab also includes public service announcements (PSAs) about voting.

X announces enhanced account security measures for government officials, candidates for office, and journalists ahead of the 2022 elections. 

Facebook updated its Ad Library to provide users with additional information about political ads, including details on targeting choices made by advertisers and how ads are delivered to different audiences. 

Twitter stopped the removal of posts related to the "Big Steal" conspiracy theory, citing that its Civic Integrity policy only applies during active election periods. The decision means that content falsely claiming the 2020 election was stolen will no longer be subject to removal.

The Facebook Papers show that based on internal studies, the company was aware of the harmful effects of their platform including spreading of false information leading up to the January 6th insurrection at the Capitol. 

Facebook announced an updated policy regarding QAnon stating, "we will remove any Facebook Pages, Groups and Instagram accounts representing QAnon, even if they contain no violent content. This is an update from the initial policy in August that removed Pages, Groups and Instagram accounts associated with QAnon when they discussed potential violence while imposing a series of restrictions to limit the reach of other Pages, Groups and Instagram accounts associated with the movement." Later updates to the policy include redirecting user searches for "QAnon" to reputable sources and working to identify other militarized social movements.

Facebook announced an initial effort to reduce the amount of political content shown in users' News Feeds, with exceptions for official government agencies and authoritative health organizations like the WHO. This shift invovled placing less emphasis on comments and shares when ranking political content and incorporating feedback from user surveys to adjust content visibility.

By March 14, 2024, Facebook decided to stop promoting any political content in users' feeds, except for accounts that users actively follow.

President Trump was banned from Facebook and Instagram due to concerns about incitement of violence.

President Trump's initial indefinite ban from Facebook and Instagram was later amended to a two-year ban following a recommendation from the Oversight Board. 

He was later allowed back on the platforms in February 2023.

Twitter introduced "Birdwatch," a community-driven program that allows users to identify and annotate tweets they believe are misleading. Users can write notes providing context to misleading tweets. Initially launched as a pilot, Birdwatch (later renamed Community Notes) requires notes to be upvoted by other members before they are publicly visible. 

Twitter has discontinued its Ad Transparency Center (ATC), which was initially designed to provide transparency for political and issue-based ads. The decision followed the platform's prohibition of political and issue ads in November 2019 and the introduction of a Cause-Based Ad certification and review process. The ATC is no longer deemed necessary given these changes, although an archived version of the tool remains accessible.

Twitter permanently suspended the @realDonaldTrump account due to the risk of further incitement of violence following the January 6 Capitol riot. The decision was based on an assessment of President Trump's tweets against the Glorification of Violence policy, which determined that his tweets violated the policy and warranted a permanent suspension.

This suspension was later lifted November 2022. 

Facebook banned the original "Stop the Steal" Facebook group within two days of its inception and after it gained more than 320,000 users. The group revolved around claims of election fraud in the U.S. presidential election. The New York Times reports that a Facebook spokesman said, "The group was organized around the delegitimization of the election process, and we saw worrying calls for violence from some members of the group."

Facebook's implements "Break the Glass" emergency measures to limit the spread of misinformation and content that could incite violence pre and post U.S. presidential election period. These measures include, in part, restrictions on groups with a history of rule violations; algorithmic changes to control virality of live or otherwise harmful content including, “demoting content that [Facebook] systems predict to be violence and incitement…[and] temporarily reducing the distribution of content [Facebook] systems predict may be false pending review by a third-party fact-checker;” and enhancing moderation by removing content violating voter interference policies.

Facebook and Twitter introduced a warning label for posts that claim a premature victory in the 2020 U.S. presidential election. One of the labels states "Official sources called this election differently." Another label says "Official sources may not have called the race when this was tweeted."

Twitter began labeling and restricting misleading Tweets, applying warnings, and using "prebunks" to provide context on issues like delayed results and mail-in voting. They also promoted Quote Tweets (encouraging users to add their thoughts when resharing) instead of Retweets. However, in December 2020, the company ended this approach, reporting that it "didn't appear to increase context."

Facebook temporarily reduced the distribution of a New York Post article about Hunter Biden while it awaited verification from third-party fact-checkers. This measure was taken as part of Facebook's policy to limit the spread of potentially false information before its accuracy could be confirmed.

Facebook and Twitter introduced a warning label for posts that claim a premature victory in the 2020 U.S. presidential election. The label reads: "Votes are being counted. The winner of the 2020 US Presidential Election has not been projected. See Election updates."

Facebook updated its policy to ban posts that "use militarized language or suggest that the goal is to intimidate, exert control, or display power over election officials or voters." This change was made to help prevent the spread of content that could incite violence or intimidation during elections.

Facebook announced that it will not run political ads for one week following the 2020 election. This decision aligns with a similar move by Google, aiming to limit the spread of misinformation and prevent potential voter manipulation during the post-election period.

Twitter restricted the visibility of a New York Post story about Hunter Biden, citing its policy on hacked materials. The platform took this action while awaiting fact-checking of the story.

To curb the spread of misinformation, Twitter implemented new design features that limit interactions with tweets labeled as misleading. Users must tap through a warning to view these tweets, and only the Quote Tweet option is available—likes, replies, and retweets are disabled. Additionally, these tweets will not be algorithmically recommended. The feature is targeted specifically at U.S. political figures, U.S.-based accounts with over 100,000 followers, and those with significant reach.

Twitter announced a policy update stating that tweets encouraging violence, or calling for interference with election results or the smooth operation of polling places, will be removed. This measure aims to ensure the integrity and safety of the election process.

Twitter introduced a "prebunking" strategy by placing information at the top of users' feeds to counter false narratives about voting by mail and election results. This initiative aimed to preemptively address misinformation before it spread widely.

Twitter implemented a policy for handling premature claims of election victories. To recognize results, the platform requires either an official announcement from state election authorities or a projection from at least two authoritative, national news outlets. Tweets with premature victory claims will be labeled and direct users to Twitter’s official U.S. election page.

Facebook implemented a policy preventing new political ads from being published during the final week leading up to Election Day. Existing ads could continue running, but no new political ads are allowed to launch. 

As part of its Civic Integrity Policy, Twitter introduced a feature prompting users to think twice before retweeting content flagged as potentially misleading. This includes a description of the Tweet or article summarizing why the content may be problematic or why the term is trending, encouraging more informed sharing decisions.

Twitter introduced heightened security measures for high-profile accounts during the 2020 U.S. election. This included enabling password reset protection, implementing more sophisticated detection systems to identify suspicious activity, and encouraging two-factor authentication for accounts of government officials, candidates for office, and journalists to protect them from hacking and interference.

The labeling policy now applies to key government officials, including foreign ministers, ambassadors, official spokespeople, and key diplomatic leaders. Twitter also announced that it will no longer amplify state-affiliated media accounts or their tweets through its recommendation systems.

The Election Hub features news in both English and Spanish, presented through Twitter Moments created by reputable outlets. It includes livestreams of debates and election events, along with voter education PSAs to inform users about voting processes and key election-related information.

Facebook expands its Dangerous Individuals and Organizations policy to include certain anarchist groups, U.S.-based militias, and QAnon, restricting their ability to organize while allowing non-violent supportive content. The change targets movements presenting "significant risks to public safety" but falling short of Facebook's full ban criteria. In September Facebook updated the policy to prohibit anyone from "running ads that paraise, support or represent militarized social movements and Qanon."

The whistleblower's documents reveal issues with Twitter's handling of election integrity. The concerns raised include the platform's inadequate measures for enforcing its election policies, highlighting gaps in its ability to manage misinformation and enforce rules effectively.

Facebook announced as part of their voting information efforts, "everyone in the US who's voting age will see a message at the top of the Facebook app with information about how to register to vote."

Also, according to CNET, Facebook began labeling posts about voting from federal officials and candidates, including presidential candidates in the United States, to direct users to credible information sources like usa.gov/voting.

Facebook introduced new ad spend trackers for U.S. House and Senate races. These trackers allow users to see how much candidates are spending on advertisements and which topics are being promoted. This feature aims to increase transparency in political advertising.

Facebook's new Voting Information Center provides authoritative information about voter eligibility, absentee ballot requests, in-person voting, and deadlines. The goal is to register four million voters, doubling the target from two million in 2016 and 2018. A link to register to vote is prominently displayed at the top of the page for all users aged 18 and older.

U.S. users can now turn off all social issue, electoral, or political ads from candidates, Super PACs, or other organizations that include the “Paid for by” political disclaimer. This option builds on Facebook's previous features for ad control.

Facebook annoucned it will begin enacting its October 2019 plan to label content from "media outlets that are wholly or partially under the editorial control of their government." It also announces that it will be labeling ads from those publishers later in the year and that some foreign state-controlled media will be blocked from advertising, stating, "State-controlled media outlets rarely advertise in the US. Nevertheless, later this summer we will begin blocking ads from these outlets in the US out of an abundance of caution to provide an extra layer of protection against various types of foreign influence in the public debate ahead of the November 2020 election in the US."

Later updates in June 2020 include blocking "adds from state-controlled media outlets targeted to people in the US;" in August 2020 adding labels on Instagram in the "About this Account section;" in September 2020 adding labels on Instagram "to posts and profiles."

In May 2020 Twitter's Civic Integrity Policy is updated to say "You may not use Twitter’s services for the purpose of manipulating or interfering in elections or other civic processes. This includes posting or sharing content that may suppress participation or mislead people about when, where, or how to participate in a civic process."

Twitter enforces rules banning content in the following categories: 1) misinformation on how to participate in the civil process, 2) suppression and intimidation, and 3) false or misleading affiliation. 

Twitter has introduced a labeling system for synthetic or manipulated media. Labels are applied based on three criteria: whether the media is synthetic or manipulated, whether it is shared in a deceptive manner, and whether it is likely to impact public safety or cause serious harm. On a case-by-case basis, Twitter may also reduce the tweet's visibility and/or alert users before they like or retweet the content.

Facebook announced a new policy to remove manipulated media, such as deepfakes and videos altered through AI, if they meet certain criteria. The policy targets media that has edited to mislead or deceive people or that might cause real-world harm. Facebook continues to work with fact-checkers and also partners with Reuters "help newsrooms worldwide to identify deepfakes and manipulated media through a free online training course."

Elections are not specifically highlighted in this version but are mentioned in future iterations. 

Facebook expanded its policies to address harmful activities related to voter and census fraud and interference. The policy prohibits content that encourages or promotes illegal participation in the voting or census process, such as offers to buy or sell votes or instructions for committing voter fraud. It also covers misrepresentations regarding the dates, locations, or methods of voting or voter registration, as well as misinformation about who can vote, candidate eligibility, or participation requirements. Additionally, the policy aims to prevent content suggesting that participating in the voting or census process may lead to law enforcement consequences, such as arrest or deportation.

Twitter announced a ban on all political content advertisements, covering both candidate ads and issue-based ads. 

Facbook announces several new policies including showing the confirmed owner of a page, labeling state-controlled media, a new U.S. presidential candidate spend tracker, clearer fact-checking labels, and $2 million investment to support media literacy projects.  

Facebook introduced "Facebook Protect," a security program designed to offer enhanced protection for the accounts of political candidates, elected officials, and their teams. The program includes additional security measures, such as two-factor authentication and real-time monitoring for suspicious activity to prevent hacking.

Facebook launched a dedicated policy section on Inauthentic Behavior, distinguishing it from its previous categorization under Misrepresentation. The policy defines Inauthentic Behavior as deceptive actions intended to conceal the identity of the organization behind a campaign, make it appear more popular or trustworthy than it is, or evade enforcement. This includes a prohibition on foreign government interference, specifically Coordinated Inauthentic Behavior conducted on behalf of foreign or government entities.

In defense of Facebook’s decision to allow political advertisements, Zuckerberg stated that banning such ads would disproportionately favor incumbents and candidates who receive more media coverage. He argued that political ads provide an essential platform for candidates, particularly those with less public exposure, to communicate their messages.

Facebook implemented a ban on paid advertisements that suggest voting is useless or that advise people not to vote. This policy shift followed a recommendation from a civil rights audit, aimed at combating voter suppression and misinformation during the election process.

Facebook introduced stricter requirements for political advertisers, aiming to confirm that the entity running the ad is U.S.-based. Organizations must now provide a Federal Election Commission ID, Tax-ID number, or a government website domain that ends in .gov or .mil. The goal is to help users on Facebook and Instagram better understand who is trying to influence them and why. Users can now view confirmed information about advertisers, including the type of identification provided, such as an EIN or FEC ID, allowing them to assess the legitimacy of the organization.

In a policy shift, Twitter decided to allow a tweet from President Donald Trump telling minority members of Congress to "go back to where you came from." This decision followed internal debates and policy reviews. 

Facebook's Ads Archive, initially launched in May 2018 to track political and policy-related ads, was replaced by the Ad Library. The new Ad Library includes all active ads across various topics, not just political or policy ads, and also displays historical data such as page creation dates, mergers, name changes, and management locations. Users can report ads for policy violations.

Responding to concerns over foreign involvement in the 2016 Election, Facebook introduced its "Inauthentic Behavior" initiative. This effort focuses on removing coordinated networks of inauthentic accounts, pages, and groups that attempt to manipulate public discourse.

Facebook announces a "war room" with a dedicated team tasked with preventing election interference and misinformation during the midterm elections. The team includes experts across the company in intelligence, computer engineering, legal matters, and researchers. 

Facebook has expanded its voter suppression policies to include new measures against misrepresentations about voting procedures. Previously, the platform prohibited offers to buy or sell votes and misrepresentations about voting dates, locations, times, and qualifications. The newly added policies now also ban false claims about voting methods, such as using online apps, and statements about whether a vote will be counted.

Twitter updated its rules to address fake accounts and misleading activities, focusing on removing accounts involved in malicious behaviors and coordinated inauthentic actions. The platform also targets the distribution of hacked materials and threats related to hacking. 

To enable external research, Twitter released a database containing tweets and account information related to Russian and Iranian efforts suspected of interfering in the 2016 elections. 

Facebook has expanded its third-party fact-checking program to cover videos and photos, in addition to text-based content. 

Twitter's #BeAVoter campaign aims to increase voter registration by promoting registration information and reminders. The initiative includes a custom emoji and a dedicated hashtag to encourage participation. Users in the United States receive prompts in their timelines with details on how to register, and the campaign highlights voter registration resources and election reminders through trending topics.

Twitter announced a political ads transparency center. This center allows users to view information about political ads, including who paid for them, their targeting criteria, and performance data. This tool was removed as of January 25, 2021.

Initially rolled out in Canada, Facebook launched a publically-searchable archive of federal-election related ads in the United States. The feature shows current and historical federal-election related ads and provide details on amounts spent, number of impressions, and demographics about the audience reached. 

Advertisers running political campaign ads for federal elections must self-identify and prove location in the United States. 

Twitter introduced candidate verification and account labeling to prevent the creation of counterfeit accounts by clearly identifying official candidates and their campaigns.

Zuckerberg testified before Congress regarding Facebook's handling of user data following the Cambridge Analytica scandal. He faced questions about the company's failure to protect users' privacy and its role in enabling the spread of disinformation, particularly during the 2016 U.S. election. Zuckerberg apologized for Facebook's lapses and outlined steps the company would take to prevent future breaches, including better data control and transparency measures. Lawmakers questioned Facebook's monopoly power, and raised concerns about regulation and oversight.

News outlets such as the New York Times and The Guardian reported that political consulting firm, Cambridge Analytica, improperly accessed data from millions of Facebook users without their consent. Whistleblower Christopher Wylie provided documents showcasing the breadth of the data usage. This data was used to create detailed voter profiles and target individuals with personalized political advertisements, particularly during the 2016 U.S. presidential election and the Brexit referendum. 

Facebook implemented a new authorization process requiring entities running ads to verify their identity and location. 

Facebook and Instagram now label political ads with a "Paid for by" disclosure to show who financed the ad. Additional information on the ad’s reach and audience is also provided.

 Twitter has prohibited advertising from Russia Today (RT) and Sputnik, both of which are state-funded media outlets associated with Russian government interests.

Twitter emphasized its election integrity efforts on combating automated accounts and manipulation networks. This focus is largely in response to concerns about the influence of Russian bots during the 2016 election.

Facebook launched a third-party fact-checking program to address misinformation. Community reports and other signals will direct stories to fact-checkers. If a story is deemed false, it will be marked as disputed, and users will be linked to an article explaining the fact-checking result. Disputed stories may also be demoted in the News Feed. Notably, politicians are exempt from this fact-checking process to ensure their statements remain publicly scrutinized and debated, whether made on their Page, in ads, or on their websites.

In 2016, Facebook and Instagram did not have specific policies addressing election-related content. Following the election, Zuckerberg publicly denied claims that fake news on Facebook had any significant impact on the election results.

Facebook allowed Donald Trump's calls for a Muslim ban to remain on the platform, even though these posts violated the company's Hate Speech policy. 

Facebook encouraged voter registrations during the 2016 election by prompting its users with a reminder to register.