Meta blames automated tools for suspending political Instagram account
Instagram Account Suspension Coincides With Early Voting Push
Automated content moderation systems suspended a major political Instagram account for hours on October 17 during an early voting push, drawing criticism from the Meta Oversight Board. Meta denied deliberate interference in the incident, attributing the disruption to mistakes in its automated tools.
-
The Tech TL;DR:
- Automated content moderation tools suspended a major political Instagram account for hours during an early voting push on October 17.
- Leaked documents and reports show automated systems struggle significantly with low-resource languages, such as deleting nonviolent Arabic content 77 percent of the time in 2020.
- The Santa Clara Principles 2.0 recommend keeping humans in the moderation loop, ensuring transparency, auditing for bias, and protecting user data.
The October 17 suspension occurred as early voting for the U.S. midterms got underway. Critics also accused Meta of boosting one candidate in Brazil’s recent presidential election. Meta denied deliberate interference in both cases, blaming automated systems that struggle to distinguish between enthusiastic follower engagement and bot attacks.
Language Disparities and Systemic Moderation Errors
Automated moderation processes high volumes of content quickly, but the systems frequently reproduce existing biases and misinterpret context. A 2025 report from the Center for Democracy and Technology found that labeled datasets in languages and dialects like Maghrebi Arabic and Kiswahili contain inconsistencies and bias due to limited hiring of native annotators. Previous leaked documents from 2020 showed that Meta’s algorithms incorrectly deleted nonviolent Arabic-language content 77 percent of the time.
These automated errors disproportionately affect journalists, activists, artists, and marginalized communities. Overreliance on automated filters has resulted in the systemic suppression of content from Palestine and the repeated misclassification of LGBTQ+ content as adult material.
Advocates Recommend Santa Clara Principles to Improve Moderation
To address these failures, advocates recommend that companies publicly commit to the Santa Clara Principles 2.0. These guidelines state that automated technologies should assist rather than replace human moderators, particularly during high-stakes periods like elections. Companies must maintain transparency regarding how automation is used in content decisions, regularly audit systems for bias in low-resource languages, and strengthen personal data protection policies.
Users require a straightforward mechanism to appeal content removals and provide context for human review. Policymakers face calls to avoid passing legislation that mandates automated moderation or dictates technical platform choices, focusing instead on public education regarding personal data collection.