Deep learning generative models have advanced recently, which has sparked worries because they can produce incredibly realistic-looking fake photographs and movies. This can cause social instability and endanger people’s integrity. In order to tackle this problem, it is imperative to create new computational models that has the ability to quickly identify fake content and warn viewers about possible photo and video alteration. This paper offers a thorough analysis of current research on deepfake content identification with deep learning techniques. By carefully examining the various false content detection categories, we hope to expand the current state-of-the-art research. In addition, we present the benefits and limitations of the reviewed studies and suggest a number of future paths to address the problems and shortfalls in deepfake detection that remain unresolved.