arXiv paper proposes CPSAINT and FRIESA-K framework for quantified residual risk in agentic AI

A new arXiv preprint in artificial intelligence, titled From Agent Failure Paths to Quantified Residual Risk: A Compositional Framework for Resilient Agentic AI, argues that agentic AI is crossing trust boundaries faster than current risk models can represent. The authors say existing approaches either describe failure mechanisms without a transferable residual-risk estimate, or produce a risk estimate while treating the internal failure path as a black box. They propose CPSAINT, a seven-layer integrity decomposition over Physical state, Sensors, Data, Compute, Actuators, Environment, and Time, paired with FRIESA-K, a residual-risk functional that maps each failure path to a quantified risk instance. FRIESA-K grounds the resistance term K in a controlled absorbing Markov model so that control effectiveness is derived from state dynamics rather than assigned as an informal score, according to the abstract. The paper reports governance observability through a separate additive penalty instead of inserting governance as a new variable in the resistance functional. The authors formalize structural composability linking valid failure paths to well-defined risk instances and illustrate the framework on two contrasting scenarios: a hard real-time warehouse robot and a governance-instrumented financial-services agent. Across both cases, they say the same layer grammar, variable semantics, and dynamic-resistance construction remain intact, yielding a compact kernel for cross-domain reasoning and quantitatively grounded composable trust.