Automated Moderation Is Here to Stay—Accountability Must Keep Pace

According to the Electronic Frontier Foundation, Meta's algorithms designed to detect terrorist content incorrectly deleted nonviolent Arabic-language posts seventy-seven percent of the time while repeatedly missing hate speech under the company's own standards. The problem persists today, worsening in languages with fewer resources where training data is inconsistent and biased. While automation scales content moderation far beyond what humans could achieve, these systems reproduce existing biases and disproportionately harm journalists, activists, and marginalized communities. The EFF proposes that automation should assist human moderators rather than replace them, companies must regularly audit for bias and ensure users can meaningfully appeal decisions, and policymakers should avoid mandating automated systems.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton