Deepfake technology, based on artificial intelligence (AI), enables the creation of realistic images, videos, and text, bringing new possibilities and posing serious ethical, social, and legal threats. The article examines key areas in which deepfakes affect contemporary society. This technology can be used to create false materials aimed at social destabilisation and propaganda, including disinformation. In the area of cyberbullying, deepfakes contribute to privacy violations such as pornography and blackmail. Additionally, generating false evidence can affect court decisions and facilitate financial fraud. This problem also affects the crisis of public trust, undermining the media’s credibility. The authors emphasise the need to introduce safeguards such as watermarking, public education and identification of AI-generated content. The need for international legal regulations and the development of technologies to detect deepfakes is indicated. The article considers ethical issues associated with this technology, minimising the risk of abuse and supporting innovation sustainably.