The emergence of autonomous artificial intelligence has confronted the traditional criminal law system, which is based on the centrality of the “spiritual element” and human fault, with a fundamental crisis of effectiveness. The inability to establish criminal intent in an agent lacking a psyche has led to the creation of a “responsibility vacuum” that threatens the rights of victims and public order. The present study, using a descriptive-analytical method and by studying library sources, addresses the main question of whether the transition from the paradigm of person-centered responsibility to a model of “risk-based responsibility” can overcome this challenge in Imami jurisprudence and Iranian criminal law. The research findings show that this transition is not only a practical necessity, but also completely defensible from a theoretical point of view. This model is rooted in solid foundations in Imami jurisprudence (including rules such as tasbib, laharr, daman yad, and the principled analysis of general knowledge) as well as the capacities of Iranian statutory law (including Article 40 of the Constitution on the prohibition of abuse of rights, Article 143 of the Islamic Penal Code on the criminal liability of legal entities, and the legislative practice in specific laws). Finally, this article argues that by shifting the focus from the “machine mind” to the “dangerous human act” in the creation and delivery of technology, a fair and efficient solution can be achieved to ensure liability in the age of artificial intelligence, thereby aligning technological innovation with social justice and security.