Die Kampf gegen Hassrede im Internet: Eine Herausforderung für die Thüringer Landesmedienanstalt

In recent years, the proliferation of hate speech online has become a significant challenge for law enforcement agencies and media authorities. In Thuringia, the Thüringer Landesmedienanstalt (TLM) has been actively combating hate speech through the use of artificial intelligence (AI) tools. Despite the successes achieved so far, the TLM faces numerous challenges in its efforts to eradicate hate speech from the internet.

The TLM’s Use of AI

In May 2021, the TLM introduced an AI-powered tool called “Kivi” to facilitate cooperation with other state media authorities and accelerate the exchange with the Federal Criminal Police Office (BKA) and the State Criminal Police Office (LKA). Kivi supports the TLM in identifying and reporting hate speech online, thereby reducing the need for manual research by the media authority. According to a spokesperson for the TLM, “Kivi helps us so that we no longer have to constantly search the internet ourselves. Our goal is to remove strafbrechende contents from the net as quickly as possible.”

The Results of the TLM’s Efforts

Between May 2021 and February 2024, the TLM received 235 reports of hate speech, including Nazi slogans, Volksverhetzung (incitement to hatred), and other strafbrechende (criminal) content. The majority of these reports were related to Facebook, with almost 90% of the reported content being removed by the platform after the TLM submitted deletion requests. In addition, Kivi can also involve the Landeskriminalamt (LKA) in the criminal prosecution of such content.

Challenges Faced by the TLM

Despite the successes achieved so far, the TLM faces significant challenges in its efforts to combat hate speech online. Not all platforms respond to deletion requests, and some platforms, such as VK and Telegram, do not remove reported content at all. This means that the TLM must continue to monitor the internet manually to identify and report hate speech. Moreover, the sheer volume of content to be reviewed is overwhelming, with Kivi currently able to analyze more than 10,000 pages per day for problematic content.

The Importance of Collaboration

To effectively combat hate speech online, collaboration between law enforcement agencies, media authorities, and social media platforms is crucial. The TLM’s use of Kivi has shown that AI-powered tools can be valuable in identifying and reporting strafbrechende content. However, the success of such efforts depends on the willingness of social media platforms to cooperate and remove reported content.

Conclusion

Combating hate speech online is a complex and ongoing challenge for law enforcement agencies and media authorities. The TLM’s use of Kivi has shown that AI-powered tools can be effective in identifying and reporting strafbrechende content. However, the success of such efforts depends on the willingness of social media platforms to cooperate and remove reported content. Collaboration between all parties involved is crucial to eradicate hate speech from the internet.