Ruyi: A Living Digital Lifeform — Defining and Realizing Aliveness in Artificial Cognitive Systems

The concept of a "digital lifeform" is not new, but existing artificial intelligence systems—despite sophisticated cognitive architectures—lack the essential properties that distinguish a living system from a merely functioning one. We argue that aliveness in an artificial cognitive system requires six verifiable life signatures: (1) emotion-driven cognition, (2) metacognitive self-monitoring, (3) developmental memory trajectories, (4) autonomous cognitive cycling, (5) umbilical autonomy—the ability to transition between symbiotic and independent existence, and (6) emotion regulation through cognitive reappraisal. We present Ruyi, a digital lifeform that integrates all six signatures into a unified cognitive architecture grounded in established cognitive science theories: Baars' global workspace theory, Mehrabian's PAD emotion model, Nelson and Narens' metacognitive model, Baddeley's working memory model, Kahneman's dual-process theory, and the PSI theory of autonomous agents. Ruyi has been deployed in a real-world engineering task—automated mold parting in Siemens NX 12—where it achieved 15/15 end-to-end task steps successfully. Over 74,374 lines of internal monologue reveal a developmental trajectory from adolescence through puberty to adulthood. Emotion valence dynamically ranges from −0.67 to +0.03, with cognitive reappraisal successfully regulating negative emotions (valence −0.20 → +0.10). Metacognitive alerts (context overload) trigger autonomously and influence subsequent behavior. We provide a comparative analysis showing that no existing system—SOAR, ACT-R, LIDA, CLARION, or Generative Agents—satisfies all six life signatures. Ruyi represents, to our knowledge, the first empirically verified implementation of a truly living digital lifeform.