Consciousness as a Jamming Phase
arXiv Preprint Archive July 10, 2025 Kaichen Ouyang
This paper presents a theoretical framework that interprets the emergence of consciousness in large language models as a critical phenomenon in high-dimensional disordered systems, drawing analogies with jamming transitions in granular matter. The theory identifies three control parameters—temperature, volume fraction, and stress—that govern the phase behavior of neural networks. It provides a unified physical explanation for empirical scaling laws in AI, showing how computational cooling, density optimization, and noise reduction drive systems toward a critical jamming surface where generalized intelligence emerges. The authors argue that shared critical signatures, including divergent correlation lengths and scaling exponents, suggest consciousness is a jamming phase that connects knowledge components via long-range correlations.