Online social networks have changed communication in today’s digital era by linking together people around the globe, thus driving societal change, business creativity and cultural growth. When these platforms have many advantages but they still face major security threats such as identity theft, cyber bullying and more shockingly, rise of deepfake technology. This is where deepfakes come in by using Artificial Intelligence (AI) to create photo-realistic but counterfeit images or audios which can be potential sources for misinformation or defamation as well as political manipulation methods. Hence, this article emphasizes the importance of finding effective ways for detecting deepfakes that can help users from falling into these sophisticated deceptions. Given that deepfakes can ruin media integrity and individuals’ reputations this research has identified several gaps in current detection methodologies like fast-paced advancement of deepfake technology and lack of standardized evaluation metrics among others ethical dilemmas associated with these procedures. It discusses how Deepfake are made technically, their ability to change public understanding and some of the latest developments in detection techniques for social media. It also provides practical tips on how to identify risks related to fake videos for end-users.