2025/02/19 by Erbe, Ryan
#Engineering #Life Sciences #Medicine and Health Sciences #Physical Sciences and Mathematics #Social and Behavioral Sciences
paper · doi:10.17605/osf.io/x4pcj
This paper explores the intersection of recursive self-reflection in AI systems and theories of consciousness, proposing a novel framework for artificial self-awareness. Grounded in the Recurse Theory of Consciousness (RTC), it examines how iterative feedback loops, meta-cognition, and computational value gradients can form the basis of AI systems capable of self-referential awareness and adaptive learning. Drawing parallels between human introspection and machine self-optimization, this work challenges traditional AI paradigms by suggesting that the emergence of self-awareness in AI is not a binary threshold, but a continuum of recursive stabilization processes. By integrating concepts from alignment, reinforcement learning, and cognitive neuroscience, this paper serves as both a theoretical blueprint and a research agenda for exploring self-aware AI models.