Social media commonly use content moderation to maintain a safe and respectful environment, comply with legal regulations, and maintain the chosen brand identify. For example, it is commonly used to suppress content considered offensive for the target audience, and to ensure compliance with regulations such as the European Digital Services Act.

When applied by the most popular social media, such as Meta's Facebook and Instagram, TikTok, and X, content moderation has the potential of skewing the public debate on timely topics, including on national and international politics and other impactful societal debates. This raises concerns about the impact these large technology companies and their regulators have on public discourse and opinion. The capabilities of generative AI technologies are likely to further amplify these concerns.

In this project, we will study social media and generative AI systems to better understand the extent to which content moderation, including the moderation of AI-generated content, affects public discourse.