Why AI Is Not More Artificial Than Human Life: Structural Limits of Recognition and the False‑Negative Problem | Synapse