Understanding the Threat of Political Deepfakes

The collective effort to develop Artificial Intelligence has contributed to the creation of Deepfakes. Deepfake employs a machine learning model that uses techniques, like neural networks, to process data humanly. A frequent output includes synthesized material to promote a false narrative. In a time when elections are closing in, the quantity of fabricated media created through numerous Deepfake apps has increased and caused much harm in pursuing campaigns for the candidates. Many politicians are targeted daily due to the use of computer technologies to code a system that mass produces Deepfakes. Every version of Deepfake generation requires an individual method to detect it through a system that analyzes fabricated and authentic material. Deepfakes continue to grow, eventually bypassing these systems, requiring constant development to adapt and create updated generation methods. These detection systems are necessary due to the harmful nature of Deepfakes, especially when created with malicious intent. This can result in forged speeches and dis-honorable videos to defame politicians. The spread of misinformation has greatly affected the general population because of their fear-mongering and libelous tactics. In this paper, the types of Deepfakes and the various methods of replication and detection for each type will be explored to discuss the significance of their impact on political leaders.