Cultural Homogenisation in Artificial Intelligence and Social Media: The Role of Annotator Labour Bias and Global Consequences

The rapid growth of artificial intelligence (AI) and social media platforms has revolutionized global communication and creativity; however, it has also amplified concerns about cultural homogenization. These technologies are largely dominated by Western companies, which infuse cultural biases into AI frameworks and the regulation of social media. This paper demonstrates that annotation labor in content moderation and Reinforcement Learning from Human Feedback (RLHF) alignment operates through distinct technical mechanisms that embed Western epistemic norms at different stages of the AI pipeline, thereby institutionalizing cultural homogenization and contributing to the silencing of non-Western views. It strives to critically consider the socio-ethical implications by lifting insights from cultural hegemony, digital colonialism, and epistemic injustice. Using examples from real life, it proposes three structural interventions: fair labor practices for data workers, context-sensitive annotation guidelines, and support for decentralized AI initiatives such as Masakhane and AI4Bharat.