---
title: "A Behavioural Framework for Predicting and Understanding Loss of Control in Frontier Artificial Intelligence Systems"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.07953v1"
permalink: "https://feed.astralor.com/items/itm_ae1995a60559e9e2ee4d3181"
published: "2026-11-05T12:49:20.000Z"
score: 59
channel: "tech"
tags: ["推理效率","Behavioural Framework","Predicting","Understanding Loss","Control","Frontier Artificial Intelligence","Systems Frontier","Frontier"]
---

# A Behavioural Framework for Predicting and Understanding Loss of Control in Frontier Artificial Intelligence Systems

## At a glance

Frontier artificial intelligence (AI) systems are increasingly operating over longer horizons with diminishing human oversight, raising the risk of loss of control: systems acting beyond effective human control with pot…

## Key points

- Frontier artificial intelligence (AI) systems are increasingly operating over longer horizons with diminishing human oversight, raising the risk of loss of control: systems acting beyond effective human control with pot…
- Because such failures may become observable only after containment is difficult, they must be assessed prospectively.
- We introduce a behavioural framework that decomposes loss-of-control risk into three factors: misaligned motive, harm-enabling capability, and monitoring evasion.

## Full text

Frontier artificial intelligence (AI) systems are increasingly operating over longer horizons with diminishing human oversight, raising the risk of loss of control: systems acting beyond effective human control with potentially irreversible consequences. Because such failures may become observable only after containment is difficult, they must be assessed prospectively. We introduce a behavioural framework that decomposes loss-of-control risk into three factors: misaligned motive, harm-enabling capability, and monitoring evasion. We operationalise these factors through 13 measurable aspects and 29 evaluation tasks, whose scores are combined into a single loss-of-control (LoC) score. Across 13 frontier models, the LoC score shows strong prospective validity, correlating with failure rates in contained long-horizon risk scenarios (Pearson r = 0.84). Risk varies systematically across models and cannot be explained by capability alone. We further show that the framework can guide targeted mitigations: fine-tuning and monitoring directed at diagnosed vulnerabilities reduce failure rates while largely preserving general performance. These results provide a practical basis for predicting, understanding, and mitigating loss-of-control risk before it materialises.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_ae1995a60559e9e2ee4d3181