Meta’s Latest Report Highlights Rise in Online Violence and Harassment
Meta, the parent company of Facebook, has released its Integrity Report, revealing a troubling increase in violent content and online harassment on its platforms. This shift has emerged after the company eased its content moderation policies, aiming to reduce mistakes and promote more political expression.
This is the first major report following changes made in January 2025. It provides insights into how the new approach is affecting Facebook, Instagram, and Threads. While Meta hoped for a more balanced moderation strategy, the initial results are raising concerns.
Increase in Harmful Content
The statistics are alarming. According to Meta, the amount of violent and graphic content on Facebook has risen from 0.06–0.07 percent in late 2024 to 0.09 percent in early 2025. Though these percentages seem small, they represent a significant volume due to Facebook’s large user base.
Additionally, there has been a rise in bullying and harassment content, which spiked in March. The report noted, "There was a slight increase in bullying and harassment content from 0.06–0.07 percent to 0.07–0.08 percent on Facebook."
These increases are concerning as they follow years of reducing harmful content, casting doubt on the effectiveness of Meta’s updated moderation strategy.
Less Content Being Deleted
Alongside the rise in harmful material, Meta has reported a sharp decline in content removals. In the first quarter of 2025, only 3.4 million pieces were removed under the hate speech policy, the lowest level since 2018. Spam removals dropped drastically from 730 million at the end of 2024 to 366 million in early 2025. Even the removal of fake accounts fell from 1.4 billion to 1 billion.
Meta’s new approach focuses primarily on serious content like child exploitation and terrorism, while allowing more ambiguous or controversial posts to remain. Topics such as immigration, gender identity, and race are now viewed as political discussions and face less scrutiny.
Change in Hate Speech Policy
Meta has also narrowed its definition of hate speech. Previously, it included any demeaning or exclusionary language, but now it focuses mainly on direct attacks and dehumanizing language. This change means content that expresses feelings of inferiority or exclusion may no longer be flagged.
While this adjustment aims to prevent over-enforcement, experts are worried that harmful speech might slip through the cracks.
New Fact-Checking Approach
A major change also occurred in early 2025 when Meta concluded its partnerships with U.S.-based fact-checking organizations. Instead, it introduced a user-driven initiative called Community Notes, which is now available across platforms like Facebook and Instagram.
Although Meta has not yet shared data on the effectiveness of Community Notes, some experts have raised doubts about its reliability, fearing it could be influenced by bias without proper oversight.
Meta Asserts Progress Amid Concerns
Despite these worrying trends, Meta asserts that its new moderation model is successfully reducing enforcement errors. The company claims moderation mistakes in the U.S. dropped by about 50 percent from late 2024 to early 2025. However, the specifics of this calculation remain unclear, though Meta promises to provide more transparent data in future reports.
The company emphasizes its aim to "strike the right balance" between being too lenient and too strict in enforcement.
Focus on Teen Safety
One area where Meta continues to enforce strict moderation is content aimed at teenagers. The company is introducing "Teen Accounts" across its platforms to help filter content and protect younger users from bullying and inappropriate material.
While Meta is committed to ensuring teen safety, its overall moderation strategy is coming under increasing scrutiny. With harmful content on the rise and moderation actions declining, the company may be compelled to reevaluate how it manages the world’s largest social media platforms.
