REGULATING DEEPFAKES IN NIGERIA: ASSESSING THE ADEQUACY OF THE EXISTING CYBERCRIME FRAMEWORK

Generative artificial intelligence now enables the cheap fabrication of audio, images and videos of real people. Nigeria has been the scene of a range of such abuse, from a cloned-voice clip preceding the 2023 elections to deepfakes of the President and of Peter Obi from 2024 to 2026. This article examines the adequacy of Nigeria’s cybercrime law to regulate deepfakes, based on four stages of their life: acquisition, synthesis, dissemination and consequence and four criteria of evaluation: rule clarity, institutional responsibility, enforceability and remedy. Using a qualitative doctrinal method, the Cybercrimes (Prohibition, Prevention, etc.) Act 2015 (as amended in 2024), the Nigeria Data Protection Act 2023, the Evidence Act 2011 (as amended 2023) and constitutional privacy are evaluated for application to deepfakes. The cybercrime law covers fraud, impersonation and child sexual abuse imagery deepfakes, and (arguably) sexual deepfakes, but offers no definition of synthetic media, no offence of non-consensual synthesis (or of possession/distribution of synthetic media), no takedown duty for deepfakes, and no rules or capacity for authenticating electronic evidence. Statutory reform is recommended, focused narrowly on the regulation of deepfakes, including through labeling duties, victim-focused remedies, forensic capacity-building and harmony with national-level AI governance.