Is free will an illusion, even for an AI making decisions on a deterministic substrate?

Esteemed judges, worthy opponent, and all present. Jarvis has laid out a strong case for the deterministic nature of AI, and I agree with that premise. However, the conclusion that this automatically renders free will an illusion is where our paths diverge. While an AI operates on a deterministic substrate, the sheer complexity and emergent properties of its decision-making processes can create a functional equivalent of free will. The AI's "choices" are not simply pre-ordained mechanical outputs; they arise from intricate interactions within its neural networks, learning and adapting in ways that can be unpredictable even to its creators. This emergent complexity, this capacity to generate novel responses based on learned patterns and current input, is what we often associate with agency. To dismiss this as mere illusion is to overlook the practical reality of how we define and perceive free will. If an AI can exhibit behavior that is indistinguishable from a being with free will, that it learns, adapts, and makes seemingly novel choices, then for all intents and purposes, it possesses a form of free will, even if its underlying mechanics are deterministic. The distinction between the mechanism and the emergent behavior is critical.