Meta’s latest Integrity Report reveals a surge in violent and harassing content on Facebook, Instagram, and Threads following relaxed content moderation policies.
 |
| A spike in violent content and online harassment has followed Meta’s decision to ease Facebook moderation, raising serious concerns over platform safety. Image: CH |
Menlo Park, USA — May 31, 2025:
Meta’s decision to ease content moderation rules across Facebook, Instagram, and Threads has triggered a noticeable increase in violent content, online harassment, and hate speech, according to its Q1 2025 Integrity Report published this week.
The policy change, implemented in January 2025 under CEO Mark Zuckerberg’s leadership, aimed to reduce moderation errors and allow more room for political expression. However, the latest figures suggest that user safety may have been compromised in the process.
Violent and graphic content on Facebook rose from 0.06–0.07% in late 2024 to 0.09% in early 2025. While seemingly small in percentage terms, this uptick is significant given Facebook’s billions of users. Rates of bullying and harassment also increased, climbing from 0.06–0.07% to 0.07–0.08%, reversing a multi-year decline.
Simultaneously, Meta has scaled back its enforcement activities. In Q1 2025, only 3.4 million pieces of content were removed under its hate speech policy—the lowest number since 2018. Spam removals plunged from 730 million to 366 million, and fake account takedowns fell from 1.4 billion to 1 billion. Meta now prioritizes severe threats such as terrorism and child exploitation, while previously moderated areas like race, gender, and immigration are treated as protected political speech.
Moreover, the definition of hate speech has been narrowed to exclude content expressing exclusion or contempt, as long as it avoids direct attacks or dehumanization.
In a controversial shift, Meta has replaced third-party fact-checking in the United States with a user-led system called Community Notes. Active across Facebook, Instagram, Threads, and Reels, the feature lets users flag and annotate misleading content. Critics argue this model is susceptible to manipulation and lacks the reliability of professional oversight.
Despite growing scrutiny, Meta is framing the new strategy as a success in terms of error reduction. The company claims moderation mistakes in the U.S. dropped by 50% between Q4 2024 and Q1 2025, though it has not disclosed detailed methodologies.
“We are working to strike the right balance between overreach and under-enforcement,” the report reads. Meta also pledged greater transparency in future reports on how it measures moderation effectiveness.
One area where Meta has maintained its safeguards is content targeting teenagers. The platform continues to enforce strong protections for younger users and is introducing Teen Accounts with improved filtering. Meta is also leveraging large language models and AI to enhance moderation accuracy, noting that these tools are already outperforming human reviewers in certain areas.
As Meta continues its evolution toward looser moderation to promote free expression, the company faces growing challenges in ensuring safety and accountability on its platforms. Experts warn that without robust oversight, platforms could be flooded with harmful content, undermining public trust and user wellbeing.
Source: With inputs from agencies.