TikTok’s latest Q1 community guidelines enforcement report indicates that the organisation removed a total of 24,954,128 videos from the platform in Pakistan. This comes as a result of videos failing to comply with content regulations, which covers everything from graphic content to hate speech or misinformation.
The data shows that TikTok had a 99.4% proactive removal rate. This demonstrates that the platform was able to remove nearly 95% of violative videos before users were able to report them, highlighting TikTok’s fast response systems.
Similarly, the platform also noted they removed the infringing content within 24 hours of the videos being uploaded or flagged. The latest data shows that 95.8% of videos were taken down within the 24-hour period.
Notably, Pakistan had the highest number of videos removed, surpassing all other countries mentioned in the report.
TikTok has had a rocky history in Pakistan. The social media platform has previously been banned four times between October 2020 to November 2021 by the Pakistan Telecommunication Authority (PTA). These restrictions have typically been voicing the same concerns over indecent and violent content that is especially damaging for the youth.
In fact, Pakistan is currently considering implementing a social media ban for all platforms, not just TikTok, for users under the age of 16, to combat the negative impacts of social media.
The recent report also provided global insights for TikTok’s content. For instance, in total, over 211 million videos were removed during Q1. Apart from videos themselves, over 73 million fake accounts were deleted. 21 million additional accounts were also deleted on the suspicion that the users were under the age of 13.
To further improve their content moderation efforts, TikTok is testing out large language models (LLMs) for their community guideline enforcement. Currently, they are conducting trials with the LLMs for comment filtering.
“We’ve started to pilot LLMs to help enforce our rules for comments on TikTok. In preliminary testing, LLM-driven comment moderation has been shown to help improve both the detection of violative comments and the accuracy of removals,” according to a company statement within the report.
Sign up for the Daily Briefing
Get the latest news and updates straight to your inbox
Network Links
GN StoreDownload our app
© Al Nisr Publishing LLC 2025. All rights reserved.