Platforms’ efforts to combat misinformation and harmful content: X and TikTok.
X (Formally Twitter)
First, X has outlined safety rules against violent, hateful, illegal, sexual, abusing contents on its website. It also has policies to protect privacy and child safety, as well as authenticity terms designed to prevent misleading or manipulating content as stated.
In 2021, X- formally twitter launched Community Notes a feature that allows users can add context to potentially misleading content, and the notes will be shown publicly, providing fact-checking by a community.
Since Elon Musk acquired twitter in 2022, X has been cut staff for safeguards and trusts. This occurred in 2023, a year in which the industry is transforming their policies post pandemic. X has faced criticism in rapid rise for Gen-Z, Neo-Nazi Antisemitism according to Combat Antisemitism Movement (CAM), a trend often attributed to Musk’s philosophy on freedom of speech.
X did not release transparency report for 2 years after take-over unlike other platforms such as Facebook and Google, finally in 2024, X released Transparency Report.
The report says, X received nearly 67 million reports of hateful conduct in the first half of 2024, and took action on more than 2,000 accounts.
According to Forbes, two major change occurred -greater reliance on AI for moderation with human review for action or flagging content, and increase in child abuse related post with less number of action taken.
While Meta also uses AI moderation, they acknowledged their algorithm do not work always properly. This raises a question of how much accountability of AI moderation has? Expert points out on the lack of mechanism is justified by the principle freedom of speech on BBC article .
Furthermore, X made it more difficult to retrieve data on posts such as comments, likes and engagement, which allows sentiment analysis and track trends. This change according to WebProNews, is intended to curb spam but also limits scrutiny.
X played significant role for political discussions surrounding US presidential election 2024. as owner of platform Musk used X to trumpet Trump policies. Center for Countering Digital Hate reported “At least 87 of Musk’s posts this year have promoted claims about the U.S. election that fact-checkers have rated as false or misleading, amassing 2 billion views”.
TikTok
Second, TikTok states community guidelines on website, which includes specific policies against misinformation and political advertisement. Contents are reviewed by an automated moderation system that scans for violations of the guidelines, if the content is flagged, it is reviewed by TikTok’s content moderators.
TikTok requires creators to label all AIGC(AI generated content). and provide reporting system for unlabeled AIGC. The platform’s AI is reportedly effective at detecting and flagging those content.
This year TikTok has introduced “Footnotes”, a tool to help users identify and report misinformation, a similar to those by Meta and X.
According to NY Times article, unlike Meta and X, TikTok is not ending any of its fact-checking programs or partnerships. TikTok says “Footnotes is not a replacement”, while X drew criticism for reducing investments in fact-check and moderation. .
To be verified, celebrities and all users have to provide government-issued ID to prove the account is genuine, like other social media platforms.
TikTok advices users to reconsider to share on fast-moving situation where the facts are not yet clear. Furthermore, TikTok also introduced new features this summer: “Creator Care” a mode that gives a filter for creators to identify offensive content before posting by one-switch and “Family Paring”, a tool that allow parents to have more oversight of their child’s account.
According to Community Guidance Enforcement report, deleted posts are 49 million in 2020, 1million deleted, Q1 2025.
TikTok partners to combat false content, with International Fact-Checking Network-accredited fact checking organizations through our Global Fact-Checking Program. In over 60 languages and 130 markets across the world.”
Conclusion
Policies and guidelines against misinformation are largely determined by companies, which means the rules are controlled by business entity. however, users do not want governmental sensor-ship either. Since the origin of social media is rooted in a pursuit of freedom of speech, we must find a balance between two principles. The automated moderation system has to keep improving to catch up fast changing environment, and effective moderation should be a result of dedicated work by both human fact-checkers and AI.
Leave a Reply