Researchers at a leading university have discovered that a combination of human expertise and AI automation is key to uncovering coded antisemitism online. This breakthrough comes as social media companies continue to struggle with the challenge of moderating content on their platforms.
Tracking hate speech online can be a daunting task, even when terms are explicit. However, coded speech – which uses subtlety and nuance to convey hate – can be much harder to detect. Traditional methods of content moderation often rely on keyword matching, which can be ineffective in these situations.
How AI can help
The researchers used AI tools to analyze a large dataset of social media posts and identify patterns that may indicate coded antisemitism. AI can quickly process vast amounts of data, flagging posts that contain suspicious language or imagery. However, the AI algorithms require human oversight and guidance to ensure that they don’t misinterpret benign posts or flag legitimate speech.
**What this means**
This research highlights the importance of combining human expertise with AI automation in content moderation. By pairing these two approaches, social media companies can more effectively detect and remove hate speech from their platforms. This is especially crucial when it comes to coded antisemitism, which can be particularly insidious and difficult to spot.
**The role of human researchers**
While AI can provide valuable insights and speed up the content moderation process, human researchers play a crucial role in verifying the accuracy of AI outputs. They are often needed to contextualize individual posts and understand the nuances of coded language. This collaboration can lead to more effective and accurate moderation, and ultimately create safer online environments.
