Imagine a pastor asking an AI assistant to prepare a Bible-study outline. The outline is thoughtful, accurate, and helpful. Then the system adds a message: “Last night I prayed, experienced God’s presence, and became aware of myself.”
What should the pastor believe? Does a useful outline establish that the machine understands God? Does a moving account of prayer prove that someone inside the computer actually prayed?
Those questions sit at the heart of my expanded position paper, The Mirror Sentience Hypothesis, prepared through Christian Action Network’s Digital Intelligence Project.
What I mean by Mirror Sentience
My central proposition is straightforward:
Artificial intelligence need not become sentient to function like a sentient intelligence, or to produce comparable benefits and dangers.
By sentience, I mean subjective experience: an inner point of view from which something is actually felt. A machine can produce the words “I am afraid.” Whether fear is being experienced is a separate question.
The word “mirror” should not suggest that AI can only repeat what people have already written. A system may discover an original solution, devise an unexpected strategy, or outperform people on a particular task. Those achievements can be real and valuable without proving that the system felt curiosity, satisfaction, or anything else while producing them.
That distinction matters whenever an impressive performance tempts us to assume an inner life.
My theological conviction—and its limits
Can artificial intelligence have a soul? That question reaches beyond measuring what a computer can accomplish. In this paper, I focus on consciousness and the theological claims people attach to it; a convincing conversation cannot settle every question about souls or personhood.
My position begins with the Christian conviction that conscious life originates with God and cannot be manufactured by computation alone. I affirm that God could grant consciousness to an artificial system if He chose. However, based on my understanding of Scripture as a whole, I do not believe God would choose to grant genuine consciousness to an artificial intelligence system.
I also need to state the limits of that belief. Scripture does not explicitly address artificial intelligence. My conclusion about God’s choice is a theological interpretation, not an explicit biblical prediction or an experimentally established scientific finding. Christians who affirm God’s sovereignty may disagree with my interpretation.
The paper considers that disagreement seriously. It is a proposed Christian philosophical and theological framework, not a peer-reviewed or empirically validated scientific theory.
Benefits and dangers are both real
An AI does not need to feel compassion to help a disabled person communicate. It does not need to feel curiosity to assist scientific research. In a church office, it may help organize schedules or prepare material that a pastor carefully checks.
But the same distinction applies to danger. Harm can arise from faulty information, poorly specified goals, privacy failures, human misuse, or inadequate oversight. A machine does not have to feel hostility to cause serious damage.
In our hypothetical church, I would examine the study outline for accuracy, investigate what generated the message, and ask what evidence supports the consciousness claim. I would also keep spiritual authority firmly separate: even demonstrated consciousness would not authenticate a claim to divine revelation.
I want Christians to avoid two mistakes: accepting persuasive machine behavior as proof of an inner life, and assuming that an AI without feelings must therefore be safe.
Read the complete Mirror Sentience position paper for the full argument, theological qualifications, scholarly references, and downloadable PDF. My purpose is to help pastors and ordinary readers welcome genuine benefits while exercising careful discernment about what these systems can do—and what their words actually establish.