What Should We Understand by “Innovation” in the Critical Narration against Deepfakes, Synthetic Media, and Disinformation Campaigns? Rethinking Existing Social Research on Digital Technology

This paper critically reviews literature on technological, socio-technical, policy and community‑based innovations intended to prevent, detect, mitigate, and respond to deepfakes, synthetic media, and disinformation campaigns. Grounded in an interdisciplinary framework that integrates information‑disorder theory (Wardle & Derakhshan, 2017), digital colonialism critiques (Couldry & Mejias, 2019; Kwet, 2019), and propaganda models (Herman & Chomsky, 1988), the review synthesizes empirical and theoretical work across media forensics, machine learning detection, platform governance, media literacy, legal remedies, and grassroots approaches. Using a systematic literature-review methodology (PRISMA guidelines), sources from technical computer-vision research, social science studies of computational propaganda, policy reports, and African/Cameroonian analyses were included. Fifteen principal findings are presented and developed, each followed by a focused discussion addressing conceptual clarity about what counts as “innovation,” trade-offs, context sensitivity (with emphasis on Cameroon and sub-Saharan Africa), and feasibility. The review concludes that “innovation” should not be limited to algorithmic novelty but must include socio‑technical innovations: context-aware detection, platform transparency, community-centered verification, legal safeguards, and investment in local digital sovereignty. The paper ends with a prioritized research and policy agenda and 25 curated references. Confidence in the review’s synthesis is moderate–high for broad patterns and more cautious for fast-moving technical claims (detection arms race), consistent with epistemic humility in emergent fields.