LATIDIA · Ciberseguridad

Ningún libro de reglas finito puede certificar todos los programas seguros: un teorema de imposibilidad para los agentes de IA de control de acceso

arXiv:2611.13367v1 Announce Type: cross Resumen: Un asistente de IA escribe pequeños programas, y un gatekeeper debe decidir, solo a partir del código, si cada uno puede ejecutarse. Supongamos que las reglas son fijas: nunca realice una

La noticia

arXiv:2611.13367v1 Announce Type: cross Abstract: An AI assistant writes small programs, and a gatekeeper must decide, from the code alone, whether each one may run. Suppose the rules are fixed: never perform a forbidden action, never let the output depend on a secret, always answer eventually. Can two rulebooks, one listing what is allowed and one listing what is not, both be complete, if every rule must be backed by evidence that a machine can check in finite time? We prove that they cannot. Checkable evidence can only establish what a finite record settles, and ``never'' and ``eventually'' are never settled by a finite record. For prohibitions the failure is