Artificial Intelligence and Hate Speech: Exposing Facebook’s Biased Algorithms and Threats to the Freedom of Expression

Abstract

Today social networks are the leading platforms for expressing opinions publicly. However, publicity on the Internet differs considerably from the generally accepted concept of publicity: when speaking any opinion or idea online, the user feels neither direct danger nor possible consequences for expressing his thoughts. Thus, Hate Speech online was created as a particular category of public statements that require special legal regulation. Today Internet platforms are actively cooperating with human rights institutions to comply with the right to freedom of expression. However, some current research shows that the activism of such companies to protect the right to freedom of expression has a massive dark side. Social networks use the soft nature of human rights regulations and the lack of clear legal definitions to violate the right to freedom of expression in their favour. They promote hate and drive toxic and harmful content for their own benefit, make deals and go for concessions with the governments, and close their eyes to the holes and failures of their own automotive systems. Therefore, this kind of conduct demonstrates an “unhealthy relationship” between human rights and Big Tech and reveals significant socio-legal gaps that have become the main focus of this study.