GOVERNABLE INTELLIGENCE: From Bounded Reasoning to Enforceable Authority (Bounded Intelligence: A Governed Intelligence Framework)
Format:
Paperback
En stock
0.47 kg
Sí
Nuevo
Amazon
USA
- Governable IntelligenceFrom Bounded Reasoning to Enforceable AuthorityIf intelligence cannot govern itself, the next question is unavoidable: what can?Governable Intelligence builds directly on the limits established in Bounded Intelligence and addresses the missing layer—authority. It examines why no amount of instruction, alignment, or internal policy can convert probabilistic reasoning into enforceable control, and why governance must exist outside the model.This book introduces an architecture-first framework for governing intelligent systems without relying on behavioral compliance or model self-restraint. It reframes refusal, constraint, and non-execution as successful outcomes, not failures.Rather than asking whether AI systems are aligned, this work asks a more fundamental question: who holds authority when an intelligent system acts—or refuses to act?Key themes include:The structural failure of model-internal governance.Why instruction is not enforcement.The separation of advisory intelligence from authoritative control.Deterministic validation versus probabilistic confidence.Governance as a system property, not a model trait.Written for architects, policymakers, and system owners responsible for AI deployment in high-risk domains, Governable Intelligence defines what it means for intelligence to be usable without being trusted.This work is a conceptual and architectural analysis. It does not describe deployable systems.
IMPORTÁ FACIL
Comprando este producto podrás descontar el IVA con tu número de RUT
NO CONSUME FRANQUICIA
Si tu carrito tiene solo libros o CD’s, no consume franquicia y podés comprar hasta U$S 1000 al año.