Synthetic media and online reality

This research focuses on the effects of advancements in "artificial intelligence" and "deep machine learning" on communication and video technologies, as well as their possible effects on society. In todays world deep learning has been successfully applied to solve various complex problems such as data analytics and computer vision. Deepfake (stemming from "deep learning" and "fake") is a type of synthetic media that has been created by home users on Reddit and similar websites. Deepfake algorithms can create fake images and videos that people cannot distinguish from their originals. These videos, created with deep learning software, emerging in the online world and seen as harmless humor, but the difference between "real" and "fake" are impossible to distinguish with untrained eyes, as they create a discussion of personali rights, however, they pose threats to privacy, democracy and national security. The purpose of this study is to explain how fake news and news images created with deepfake algorithms for the purpose of creating public opinion on mainstream media, social media and internet news sites for manipulating social perception, and how the social construction of reality with media images came true with examples and sources. In the research, the working principles of computer, graphic card and deep learning algorithms from 2017 to today, as well as the effects of communication technologies and media on society and people were investigated.