Deepfake technology has transformed the production and synthesis of digital media by exploiting the advanced technologies of artificial intelligence and machine learning. Using generative adversarial networks (GANs) and associated methods, deepfake allows for the creation of incredibly lifelike images, movies, and sounds that are frequently indistinguishable from real-world material. This further gives rise to serious ethical, legal, and societal issues even if it has significant role in fields like accessibility, education, and entertainment. The misuse of deepfake in compaigns, cyber threats, non-consensual videos, political concerns hamper the individual’s privacy and their public trust. This paper introduces the deepfake with its working, explore the approaches utilized to create and detect the deepfake. The paper further examines the ethical, legal concerns and their psychological social impact on deepfake with its positive and negative use cases in different cross-domain application areas. It also studies some of the notable case studies on deepfake with the regulatory compliance implications. Finally, different deepfake threats and their countermeasures are also detailed with future research directions.