Russian interference in elections

Russian interference in democratic elections refers to state-sponsored disinformation and manipulation campaigns designed to influence electoral outcomes and undermine trust in democratic institutions. The primary vehicle is the Internet Research Agency (IRA), a St. Petersburg-based troll farm that operated large-scale coordinated inauthentic behavior campaigns across social media platforms targeting U.S., European, and other elections.

Key operations

2016 U.S. Presidential Election: IRA operated 1,000+ fake accounts across Twitter, Facebook, Instagram, and Reddit with budgets exceeding $1M/month. Campaigns amplified divisive content on both sides of polarizing issues (race relations, gun control, immigration, LGBT rights) to deepen societal divides. Targeted amplification of negative coverage of Hillary Clinton; some support for Bernie Sanders and Jill Stein to fragment Democratic opposition.

2020 U.S. Presidential Election: Continued IRA operations but with reduced reach due to platform enforcement.

European elections: Documented interference in UK Brexit referendum, French presidential elections, and German elections.

Methods and tactics

Troll farm operations: Human operators managing large portfolios of fake accounts
Bot amplification: Automated accounts coordinating to amplify IRA content
Divisive content: Strategic amplification of inflammatory narratives on both sides to maximize polarization
Cross-platform coordination: Coordinated activity across multiple platforms for force-multiplication
Targeting: Micro-segmentation and audience-specific messaging tailored to political preferences and vulnerabilities
Real-world coordination: Integration with offline events and hack-and-leak operations

Academic documentation

Research has established:

Conservative Twitter users amplified Russian troll content 31× more frequently than liberals during 2016 election
Russian trolls deployed a pro-Trump, conservative agenda
Bots representing ~5–6% of users engaging with Russian trolls
Geographic concentration of engagement in Southern states (Tennessee, Texas)

Impact and implications

Information pollution: Documented spillover of state-sponsored disinformation into mainstream media
Polarization amplification: Asymmetric amplification of divisive content on both sides deepened political divisions
Platform vulnerability: Demonstrated large-scale exploitation of social media platforms' architecture and business models
Democratic erosion: Fueled post-election narratives of "rigged system" and "stolen election"

Key papers in this wiki

2019-im-russian-troll-accounts — Machine learning models trained on known IRA troll accounts achieve 78.5% precision detecting trolls; when applied to accounts mentioning journalists in 2018, flags 3.7% as likely trolls with ~70% human validation rate; shows trolls remain active but are not primarily bots
2018-badawy-russian-interference-twitter — Empirical analysis of Russian IRA campaigns' 2016 election effects; shows conservative users amplified trolls 31× more, produced 36× more amplifying tweets
2020-bail-russian-ira-polarization — Longitudinal quasi-experimental evidence that Russian propaganda directed at already-polarized audiences produces minimal attitude change