Gender bias in machine translation (MT) systems poses a significant challenge to achieving accurate and inclusive translations. This paper examines gender bias in machine translation for Telugu and Kannada, two major languages of the Dravidian language family, focusing on the impact of gender inflections on translation accuracy. Using Google Translate and ChatGPT, it explores how Chain of Thought (CoT) processing mitigates bias, reducing it from 80% to 4% in Telugu and from 40% to 0% in Kannada. The findings highlight the importance of using strategies tailored to each language to ensure fairness in data preparation and machine translation output.