The rapidly growing digital sphere, artificial intelligence (AI) is reshaping the politics in Pakistan. One of the most striking developments in this context is the rise of deep fake generated videos and audios which are blurring the line between fact and fabrication, leading to the questions about truth, trust, and the health of democratic debate. This study explores how deep fakes are emerging as a new instrument in Pakistan’s political landscape. It manipulates and damage reputation that is hard to undo. The media, which is already highly polarized, deep fakes in the social media platforms, find an audience that is quick to react, but often slow to verify. Factors such as uneven digital literacy, weak fact-checking infrastructure and the absence of comprehensive legal safeguards make the personalities and institutions vulnerable to AI- driven misinformation. At the same time, the competitive and often adversarial nature of party politics encourages the tactical use of such tools. This research argues that it is also deeply political and social AI driven phenomena. Countering deepfakeswill require more than just advanced detection software; it demands coordinated efforts among lawmakers, civil society, the tech sector and the public. Education in media literacy, stronger regulatory frameworks and ethical commitments from digital platforms are all part of the solution. By examining deepfakes through the lens of Pakistan’s digital political life, this study aims to contribute to broader debates about AI’s role in transforming democratic spaces. The challenge is not just to expose fabricated truths, but to preserve the fragile trust on which meaningful political discourse depends.