Human rights considerations in navigating AI content moderation : safeguarding freedom of expression in the age of social media

Abstract

This dissertation explores the rapidly evolving field of Artificial Intelligence, focusing on AI-driven content moderation tools, their role on social media platforms, and their implications for freedom of expression. Despite being a relatively recent area of research, AI has already significantly reshaped how users interact with content and express opinions online. Its growing influence raises critical questions about transparency, accountability, and the delicate balance between safeguarding online spaces and protecting fundamental rights like free speech. This study commences by introducing concepts of Artificial Intelligence, The Media and Freedom of Expression according to Article 10 of the ECHR, allowing readers to grasp main concepts and familiarise themselves with the topic at hand. This work then delves deeper into the role played by social media intermediaries in the context of freedom of expression. Here, case law and other legislative tools are examined to understand the roles and responsibilities held by social media companies, ultimately uncovering the necessity of AI content moderation tools in fulfilling platforms roles of a ‘purveyor of information’ as well as a watchdog within society. Through this discovery, the negative impacts of AI content moderation tools on freedom of expression are also uncovered and further reinforced in a case study on Facebook, ultimately used to showcase whether or not that discussed through theory reflects the reality of AI content moderation tools in social media. Whilst social media platforms maintain an element of responsibility with regards to content posted and shared online, other players are also crucial in ensuring that freedom of expression is respected throughout. Here, an analysis of legislative frameworks and their effectiveness in ensuring a balance between AI tools and protection of fundamental human rights is discussed. The study ultimately questions the existing implications of AI driven content moderation tools on freedom of expression and whether existing legal frameworks address these challenges produced through the introduction of artificial intelligence. This work suggests that whilst content moderation tools are essential in maintaining a safe environment online, the existing drawbacks within artificial intelligence itself make it necessary to maintain human involvement in automated content moderation.