Abstract
This article examines the legal risks that deepfakes pose to honour, dignity, privacy, the right to one’s image,business reputation and public safety in the Republic of Kazakhstan. The problem is framed as a new means ofharming individuals and public order through synthetic audio, photographic and video content, rather than anisolated technological defect. Sources comprise Kazakhstan’s Constitution, Civil and Criminal Codes, Code ofAdministrative Offences, the Laws “On Artificial Intelligence” and “On Online Platforms and Online Advertising”,international instruments, and EU, US and UK legal approaches. Formal legal, comparative, systemic and doctrinalmethods and legal modelling are employed. Existing safeguards are fragmented: addressing defamation, privacyviolations, fraud and dissemination of false information, they do not fully encompass synthetic impersonation. Asequential framework from detection to restoring victims’ rights is proposed, with limits on using detection resultsas digital evidence. The author proposes a differentiated model: civil remedies for urgently halting disseminationand removing harmful deepfakes; priority application of existing offences and development of targeted criminal-lawcriteria for non-consensual intimate deepfakes, using minors’ images in sexualised contexts, fraud, extortion andimpersonation seriously threatening public safety; and specific platform obligations without duplicating existinglabelling requirements.