Startup Uses AI to Prevent Hate Speech and Spam Online

© Conversario

The Berlin-based startup Conversario uses artificial intelligence to protect and understand digital communities. Their tools limit hate speech and spam, allowing more room for real dialogue to happen online.

By monitoring 100% of comments on a group’s website or social media profiles, Conversario provides 24/7 protection against unwanted speech and trolling users. The software fosters a positive atmosphere and genuine dialogue online, keeping constructive users on a site longer and ensuring they return.

Conversario’s Moderation AI vets all incoming comments using an algorithm that involves a massive “black and gray list,” a set of heuristics and a trained corpus based on millions of assessed comments.

The algorithm distinguishes comments into five categories: very toxic, toxic, spam, to review and positive. A community manager stays in full control as they decide what automated action Conversario should process based on the analysis results. That ranges from fully-automated comment approval to full-manual review by the Community Manager. Conversario’s Moderation AI continually keeps learning by the manual remoderation decisions.

In 2019, Conversario aims to grow the software from Comment Monitoring AI to Community AI, exploring how AI can support an efficient development of digital communities in all stages. Keep an eye on Conversario to follow these developments!