THE IMPACT OF POLITICAL DEEPFAKE CONTENT ON DECLINING PUBLIC TRUST: A QUANTITATIVE ANALYSIS OF FIRST-TIME VOTERS ON SOCIAL MEDIA

Abstrak

The evolution of artificial intelligence has given rise to the deepfake phenomenon, a deep learning algorithm-based audio-visual manipulation capable of producing highly realistic political disinformation. This study aims to quantitatively examine the impact of exposure to political deepfake content on social media on the decline of public trust, with a specific focus on the segment of first -time voters. Based on the theoretical foundation of Information Processing Theory and the concept of Liar's Dividend, first-time voters are identified as the most vulnerable group and active consumers of digital political information (Chesney & Citron, 2019). This study uses a quantitative approach with an explanatory survey method on a sample of first-time voters spread across several urban areas. Primary data were collected through a structured Likert-scale questionnaire and analyzed using the Structural Equation Modeling (SEM) method based on Partial Least Squares (PLS). The results are expected to show that the frequency of exposure and the perceived realism of political deepfake content have a significant negative influence on first-time voters' trust in political institutions and the integrity of the general election process. The manipulative nature of content not only degrades trust in specific political figures but also fuels systemic skepticism that undermines democratic legitimacy (Vaccari & Chadwick, 2020). These findings provide an important contribution to the growing literature on digital disinformation and urge policymakers to formulate new regulatory and media literacy frameworks to protect the integrity of young voters in the attention economy.