TikTok Dismisses 150 Moderators in Berlin, Replacing Them with AI and Subcontractors
berlijn, zaterdag, 2 augustus 2025.
TikTok is set to dismiss 150 content moderators in Berlin, including Dutch-speaking employees. These moderators, responsible for monitoring harmful content, will be replaced by artificial intelligence and subcontractors. Unions and employees fear that this decision will make the platform less safe, as AI struggles to detect subtle forms of hate speech and fake news. Moderators have already noted that they intervene too late on dangerous trends due to understaffing. The European Commission has launched an investigation into TikTok’s AI moderation practices.
TikTok Dismisses 150 Moderators in Berlin, Replacing Them with AI and Subcontractors
TikTok is set to dismiss 150 content moderators in Berlin, including Dutch-speaking employees. These moderators, responsible for monitoring harmful content, will be replaced by artificial intelligence and subcontractors. Unions and employees fear that this decision will make the platform less safe, as AI struggles to detect subtle forms of hate speech and fake news. Moderators have already noted that they intervene too late on dangerous trends due to understaffing. The European Commission has launched an investigation into TikTok’s AI moderation practices [1][2][3].
Impact on Dutch-Speaking Moderators
The Dutch-speaking moderators working in Berlin are also part of this group of 150 employees. They are responsible for reviewing videos for harmful content, including hate speech, fake news, and sexually explicit videos. According to an anonymous Flemish content moderator who has worked for TikTok in Berlin for 3.5 years, interventions on dangerous trends are often too late due to understaffing or lack of communication from management. Replacing us with artificial intelligence will make the app less safe, the moderator asserts [2].
Unions and Employees Take Action
The German union Ver.di confirms TikTok’s decision and is already in talks with the company to prevent the layoffs. Employees in Berlin have already staged two strikes against the decision, the first ever in Germany against a social media platform. TikTok is currently taking legal steps to establish a mediation committee, as required by German law [1][2][3].
AI and the Challenges of Content Moderation
AI expert Olivier Cauberghs points out that AI is a useful tool but is not sufficient to detect all subtle forms of hate speech, such as those conveyed through emojis, code words, and music. He fears that more inappropriate and illegal content, such as hate speech, fake news, and sexually explicit videos, will circulate on the platform. Cauberghs sees a dangerous precedent, as we have seen only reports in recent years of social media companies reducing their moderation teams [2].
European Commission Launches Investigation
The European Commission has launched an investigation into TikTok’s AI moderation practices, particularly following the layoffs in Amsterdam last year. The investigation is ongoing and will examine how TikTok ensures the safety of its platform. TikTok has indicated that nearly 90% of videos removed for violent content are deleted based on AI. However, the company must demonstrate that it is deploying sufficient resources to keep the platform safe, as required by law [1][2][3].
Practical Tips to Identify Fake News
To combat the spread of fake news, we offer some practical tips for readers to identify fake news:
- Check the Source: Ensure the news comes from a reliable and well-known outlet.
- Consult Multiple Sources: Compare information with multiple reliable sources to get a nuanced view.
- Fact-Check: Use fact-checking websites to verify claims.
- Language and Style: Look out for unprofessional language, typos, or sensational headlines that seem designed to shock or manipulate.
- Date and Context: Check the date and context of the news to see if it is current and consistent with other information.
- Read Comments and Feedback: Read comments and feedback from other readers to hear different perspectives.
By following these steps, readers can better assess whether news is reliable and help reduce the spread of fake news [GPT].