This survey paper introduces a proactive, lifecycle-based approach to detecting inauthentic digital content, moving beyond traditional reactive methods. It synthesizes research from machine learning and social science, using the C5 Interaction Model to analyze the creation, seeding, and propagation of synthetic narratives. The paper reviews state-of-the-art techniques for modeling coordinated inauthentic behavior and proposes future research directions focused on building anticipatory and resilient systems against rapidly evolving GenAI threats. AI
IMPACT Provides a framework for developing more resilient digital ecosystems against AI-generated disinformation.
RANK_REASON Academic survey paper on AI safety and detection methods. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →