A study by researchers from Ruhr University Bochum, the GESIS Leibniz Institute, and the CISPA Helmholtz Center reveals that AI-generated profile pictures are increasingly utilized for political manipulation on the platform X. Over half of the analyzed accounts were created in 2023, often in bulk, a strategy typical of disinformation campaigns. This aligns with findings from the Center for Countering Digital Hate indicating that Elon Musk's political posts have garnered significant attention, many containing misleading claims about U.S. elections. The study found synthetic-face accounts had distinct patterns, such as fewer followers and lower engagement levels compared to genuine profiles, indicating their orchestrated nature. Additionally, these fake accounts frequently engage in spamming and promote controversial topics, notably political issues and crypto scams. The researchers aim to enhance methods for detecting various generative AI images to combat coordinated inauthentic behavior effectively.

Source 🔗