The utilization of artificial intelligence capabilities in filmmaking has led to technical and content-related transformations within the industry. One of the prominent tools in this area is deepfake technology, which leverages deep learning algorithms to enable the realistic recreation and alteration of faces and voices. This research aims to identify the opportunities and challenges of deepfake technology in filmmaking, and seeks to provide a set of ethical requirements for its use. The research method is a systematic review of data, and the information gathering technique is analytical note-taking. The data processing tool is King's (1998) thematic coding. The research articles were selected using purposive sampling. Reliability was assessed through the constant comparative method. The findings indicate that the most significant potentials of using deepfakes in filmmaking are improving visual effects, content production and creation of virtual characters, enhancing video quality, innovation in screenwriting, reducing costs, and increasing creativity. The most significant challenges are privacy violations, identity theft, misleading audiences through fake news and political misuse, threatening employment in the film industry, and blurring the lines between reality and fiction. Furthermore, the findings indicate that the ethical requirements for using deepfakes in filmmaking should include avoiding algorithmic bias, respecting individual privacy and rights, upholding justice and non-discrimination, adhering to AI ethics, committing to transparency and data accuracy, raising public awareness, observing ethical and Islamic ethical standards, taking responsibility for the consequences of using AI, and preventing the dissemination of misinformation.