Social media affects voting and elections in five concrete ways: it drives turnout through peer influence, lets campaigns target voters individually with tailored messages, narrows what any one person sees through engagement-driven algorithms, accelerates the spread of false claims and foreign influence operations, and operates inside a legal framework that hasn’t fully caught up with the technology. Each of these has been measured, litigated, or regulated, and each changes something about how you experience an election.
Turnout: The Measurable Push to Vote
The clearest evidence that social media moves votes comes from a Facebook experiment during the 2010 midterm elections. The platform showed 61 million users an election-day message displaying which of their friends had already voted. Researchers estimated the message generated about 60,000 votes directly and another 280,000 through social contagion, for a total of roughly 340,000 additional votes — about 0.14% of the voting-age population. The entire effect came from close friendships. Messages from acquaintances or from the platform itself produced no measurable change in behavior.1National Library of Medicine. A 61-Million-Person Experiment in Social Influence and Political Mobilization
That finding explains why campaigns and nonprofit groups now rely so heavily on shared graphics with polling-place information, registration deadline reminders, and coordinated posts around early voting. A friend’s link carries weight a public service announcement doesn’t. For grassroots organizations without television budgets, this peer effect is often the entire mobilization strategy.
Microtargeting: Different Ads for Different Voters
Campaigns combine voter registration rolls, consumer purchase records, browsing history, and social media activity to build profiles of individual voters. Microtargeting then lets a campaign show different messages to different people based on predicted concerns. A voter tagged as a likely suburban parent might see an ad about school funding, while a young urban renter sees one about housing costs, from the same candidate on the same day.
The upside is relevance. The downside is that different audiences may never see each other’s ads, which makes it harder to notice when a campaign says contradictory things to different groups. Meta partially addresses this with its Ad Library, a public database that archives all political and social-issue ads, including targeting details, spend amounts, and estimated reach, for seven years.2Meta. Meta Ad Library Tools Other platforms disclose less, so much of the microtargeted ecosystem stays out of public view.
Why Your Feed Narrows
Platform algorithms optimize for engagement, and content that provokes strong reactions performs well by that measure. Over time, your feed reflects more of what you already believe and less of what challenges it. Compromise positions get less visibility because they generate less engagement. Candidates who take extreme stances earn outsized attention.
This isn’t a deliberate scheme. It’s a structural incentive: the business model rewards engagement, strong emotions drive engagement, and politically charged content reliably produces strong emotions. The practical effect on voting is that issues start to feel more binary than they are, and voters inside tightly curated feeds may rarely encounter the reasoning behind opposing views.
False Claims, Foreign Interference, and AI Fakes
Two categories of bad information move through social feeds during elections. Misinformation is inaccurate content shared by people who believe it. Disinformation is fabricated to deceive: fake screenshots of ballot counts, invented candidate quotes, or false claims about voter eligibility designed to keep people home.
Foreign governments exploit this environment systematically. The FBI and CISA have documented tactics used by Russia and Iran during recent U.S. election cycles, including fake news websites designed to mimic outlets like the Washington Post and Fox News, and paid social media influencers spreading divisive content without disclosing the foreign source.3CISA. FBI and CISA Issue Public Service Announcement Warning of Tactics Foreign Threat Actors Are Using The goal is often not to elect a specific candidate but to deepen divisions and erode trust in the process.
AI tools have made this cheaper and faster. Realistic fake video and audio of candidates can now be produced quickly, and a convincing deepfake released in the final days before an election can shift votes before it’s debunked. Corrections rarely travel as far as the original.
Federal regulation of AI in political content is limited. The FCC proposed a rule in August 2024 that would require broadcasters to disclose AI-generated content in political ads, but it targets television and radio, not social media platforms, and had not been finalized as of its publication.4Federal Register. Disclosure and Transparency of Artificial Intelligence-Generated Content in Political Advertisements The FEC has considered whether existing campaign fraud rules cover deceptive AI content but has not issued a final ruling.
States have moved faster. At least 15 have enacted laws addressing AI deepfakes in election content, with varying approaches. Arizona requires clear disclosure on any AI-generated content depicting a candidate distributed within 90 days of an election. Florida and New Mexico impose both civil and criminal penalties for distributing deceptive AI media about candidates.5National Conference of State Legislatures. Deceptive Audio or Visual Media (Deepfakes) 2024 Legislation Other enacted states include California, Colorado, Michigan, New York, and Wisconsin. Your protections depend on where you live, and none of these state laws reach a foreign actor posting from overseas.
The Rules Behind Political Ads You See Online
Disclaimers on Paid Political Content
Since March 2023, FEC rules require any paid political communication placed on another person’s website, app, or advertising platform to carry a disclaimer identifying who paid for it. The disclaimer must be readable without clicking or scrolling, large enough to read, and displayed with adequate color contrast. Video ads must show it for at least four seconds.6Federal Register. Internet Communication Disclaimers and Definition of Public Communication When the format is too small, an adapted disclaimer can direct viewers to the full information within one click.
A significant gap: if a campaign pays an influencer to post from their own account without boosting the post as a paid ad, current FEC rules do not require a political disclaimer. Disclosure kicks in only when the platform itself is paid to push the post. Followers may have no way of knowing a post was sponsored unless the influencer chooses to say so.
The Foreign Spending Ban
Federal law prohibits foreign nationals from spending money to influence any federal, state, or local election, including paying for digital ads, making campaign contributions, or funding communications that advocate for or against a candidate. Anyone who knowingly solicits or accepts such spending also faces liability.7Office of the Law Revision Counsel. 52 USC 30121 – Contributions and Donations by Foreign Nationals The implementing regulation extends the ban to all election-related disbursements, including payments for online electioneering communications.8eCFR. 11 CFR 110.20 – Prohibition on Contributions, Donations, Expenditures, Independent Expenditures, and Disbursements by Foreign Nationals
Enforcement is the hard part. Foreign actors can route spending through shell companies, cryptocurrency, or domestic intermediaries. The Honest Ads Act, which would require large digital platforms with at least 50 million monthly visitors to maintain public files of political ad purchases, including copies of the ads, targeting data, and buyer contact information, has been introduced in multiple sessions of Congress but has not advanced past committee.9Congress.gov. S.486 – Honest Ads Act
Section 230 and Platform Discretion
Section 230 of the Communications Decency Act provides that platforms are not treated as the publisher of content their users post, so they generally aren’t liable for political misinformation spread by individuals. The same provision protects a platform’s decision to remove or label content it considers misleading. The rules governing what appears in your political feed are largely set by private companies responding to public pressure, advertiser concerns, and their own engagement calculations.
When Online Behavior Crosses Into Illegal Voter Suppression
Intimidating or threatening someone to prevent them from voting is a federal crime, and the law doesn’t distinguish between in-person and online conduct. Under 18 U.S.C. § 594, anyone who intimidates, threatens, or coerces another person to interfere with their right to vote in a federal election faces up to one year in prison, a fine, or both.10Office of the Law Revision Counsel. 18 USC 594 – Intimidation of Voters A threatening direct message aimed at keeping someone from the polls falls within the statute’s reach.
Fabricated information about when, where, or how to vote is a related tactic. Posts falsely claiming an election has been rescheduled, that certain voters need special documentation, or that you can vote by text message are designed to suppress turnout. Proposed federal legislation would explicitly criminalize this kind of election-related deception, but it has not been enacted, and coverage under existing law depends on whether the conduct amounts to intimidation.
If you encounter suspected voter intimidation or suppression online, you can report it to the Department of Justice’s Civil Rights Division at civilrights.justice.gov/report or by calling 1-855-856-1247. Threats of violence should go to local police first, then to your nearest U.S. Attorney’s Office or FBI field office.11Department of Justice. Voting Resources