The rapid development of artificial intelligence (AI) has brought significant transformations across various sectors, yet it also poses new challenges to the protection of human rights (HAM) in the digital era. This article analyzes the criminal law challenges in Indonesia concerning the misuse of AI, focusing on AI's potential as a legal subject and models of criminal liability under the new Criminal Code (KUHP, Law No. 1 of 2023) and the Electronic Information and Transactions Law (UU ITE). This research indicates that the current Indonesian criminal justice system, predominantly based on the concept of human and corporate legal subjects, faces difficulties in accommodating criminal liability for actions wholly or largely caused by AI. The absence of dedicated AI regulations in Indonesia leads to reliance on existing regulations, such as UU ITE and the Personal Data Protection Law, which are often inadequate. The article identifies legal gaps in the KUHP and UU ITE concerning the definition of legal subjects and the attribution of fault for AI. Several potential models of criminal liability are proposed, including the artificial personhood approach, producer/developer-based liability, operator/user-based liability, and the application of strict liability principles and audit standards. The urgency of cybercrime law reform in Indonesia is emphasized to ensure a comprehensive legal framework that protects human rights from AI misuse.