Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment

In three linesPosition paper arguing that a three-layer probabilistic architecture (semantic intent/policy compliance, environmental validity, dynamical feasibility) is structurally required for safe LLM agent deployment. Each layer must independently certify one safety dimension via composable probabilistic guarantees.

Summary generated by Claude — human-verified