The Deterministic Horizon: Impossibility Results as Design Specifications for Trustworthy AI Systems

This thesis redefines impossibility results as design rules for trustworthy AI, demonstrating an accuracy ceiling determined solely by architectural limitations. It introduces the