123RF

Streamlining Content Screening from Manual to Automated with AI

123RF faced challenges in its content moderation process, relying heavily on manual screening, which was time-consuming and prone to errors. To address this, the organization implemented Anthropic Claude 3 models hosted on Amazon Bedrock, transitioning to an automated system that significantly enhances the accuracy and speed of image screening. As a result, content screening speeds increased by tenfold compared to human efforts, and batch translation costs plummeted from $120,000 to just $6,000 for nine languages. This transformation not only allows for the rapid flagging of inappropriate content within 1-2 seconds but also streamlines model management, ultimately improving user experience and expanding market reach.

This case study is in our research queue. We publish each one with the outcomes and the sources behind them, which takes time — check back, or read the ones that are ready.