As Generative AI tools becomes more integrated into our daily lives, the ethical implications of the data used to train these models have come under intense scrutiny. Bias in training datasets can lead to skewed results that reinforce stereotypes or discriminate against specific demographics. In this post, we explore the responsibility of developers to curate diverse datasets and implement fairness constraints during the training phase. We also look at real-world examples of algorithmic bias and discuss the emerging frameworks designed to ensure transparency and accountability in AI development.