Creative Power, Deceptive Potential: How Image Generators are Reshaping Visual Truth

Abstract

The rapid evolution of generative AI image models, such as Midjourney, DALL·E, and Google Gemini, has transformed visual communication by democratizing creativity while simultaneously amplifying risks of disinformation. This paper examines the dual nature of these technologies, highlighting their capacity to enhance artistic expression, accelerate creative workflows, and foster new forms of participation, while also enabling large-scale synthetic deception with political, social, and economic consequences. Through the analysis of case studies, industry practices, and regulatory frameworks, the study outlines both opportunities and threats generated by AI image tools.