NUWA Risk Governance · Study
Async control: Stress-testing asynchronous control measures for LLM agents
NUWA research note · independent review pending
NUWA Research Note
Research questionDesign and assumptionsSource findingsNUWA interpretationLimits and contrary evidenceResearch connectionsResearch and decision relevanceInstitutions and rolesOriginal resourcesModel-assisted reading · Selected sections of the original full text or official article; a complete section-by-section audit remains pending.
At a glance
After-the-fact ensemble monitors performed well in a bounded red-blue game; deployment risk still depends on system design.
Research question
Can asynchronous review detect sabotage soon enough to support a credible deployment-risk argument?
Design and assumptions
A prompt-constrained attack policy and evolving monitors compete across software environments with held-out and OOD tests; a separate model extrapolates detection to hypothetical deployment.
Source findings
The strongest ensemble outperforms the best single monitor under the reported calibration.
Delay before harmful effect and requiring multiple successful actions can improve modeled safety more than monitor accuracy alone.
NUWA interpretation
NUWA interpretation: empirical detection and simulated deployment safety are different layers of evidence.
Limits and contrary evidence
Assumptions include the studied attack policy, legible reasoning access, no collusion or monitor jailbreaks, and a hypothetical deployment model; subtle OOD sabotage is harder.
Research connections
NUWA connection: practical-control-monitoring formalizes latency and recovery requirements; BashArena uses a different synchronous metric and cannot supply matched rates.
Research and decision relevance
NUWA recommendation: preserve the red-team restriction and calibration data, then stress-test deployment assumptions and recovery delays separately.
Precise source locations
- Async control: Stress-testing asynchronous control measures for LLM agents · 2 SettingsSelected original resource https://arxiv.org/html/2512.13526; sha256:1b1574067ef8c25c9f507be0454732520754e71c9624fc7b27a3e9aeabb978bc; archived source capture; exact capture time not independently established
- Async control: Stress-testing asynchronous control measures for LLM agents · 3 Red-Blue Team GameSelected original resource https://arxiv.org/html/2512.13526; sha256:1b1574067ef8c25c9f507be0454732520754e71c9624fc7b27a3e9aeabb978bc; archived source capture; exact capture time not independently established
- Async control: Stress-testing asynchronous control measures for LLM agents · 4 ResultsSelected original resource https://arxiv.org/html/2512.13526; sha256:1b1574067ef8c25c9f507be0454732520754e71c9624fc7b27a3e9aeabb978bc; archived source capture; exact capture time not independently established
- Async control: Stress-testing asynchronous control measures for LLM agents · 5 Estimating risk across a deploymentSelected original resource https://arxiv.org/html/2512.13526; sha256:1b1574067ef8c25c9f507be0454732520754e71c9624fc7b27a3e9aeabb978bc; archived source capture; exact capture time not independently established
- Async control: Stress-testing asynchronous control measures for LLM agents · 5.2 Model Uncertainty and LimitationsSelected original resource https://arxiv.org/html/2512.13526; sha256:1b1574067ef8c25c9f507be0454732520754e71c9624fc7b27a3e9aeabb978bc; archived source capture; exact capture time not independently established
- Async control: Stress-testing asynchronous control measures for LLM agents · 6 DiscussionSelected original resource https://arxiv.org/html/2512.13526; sha256:1b1574067ef8c25c9f507be0454732520754e71c9624fc7b27a3e9aeabb978bc; archived source capture; exact capture time not independently established
Source checking, extraction, editorial approval and replication are recorded separately. Methodology →
Institutions and roles
Cite and reuse
Cite the original publication for its findings. Cite this NUWA record separately for its explanation and retained snapshot.
Original publication · BibTeX
@misc{async-control,
title = {Async control: Stress-testing asynchronous control measures for LLM agents},
url = {https://www.aisi.gov.uk/research/async-control-stress-testing-asynchronous-control-measures-for-llm-agents},
year = {2025},
date = {2025-12-15},
version = {Source capture 2026-10-25},
note = {Source capture 2026-10-25}
}
NUWA record · BibTeX
@misc{async-control-nuwa,
title = {Async control: Stress-testing asynchronous control measures for LLM agents — NUWA research record},
url = {https://whitzard.tech/en/nuwa/whitzard-index/work/async-control/},
version = {2026-10-29.1},
organization = {NUWA Frontier AI Safety Lab},
howpublished = {Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--async-control.json},
note = {NUWA research note. Source-reading scope and independent review status are stated on the record. Snapshot 2026-10-29.1. Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--async-control.json.}
}
Evidence snapshot 2026-10-29.1 · Index updates · RSS
References
- [1] Async control: Stress-testing asynchronous control measures for LLM agents. 2025-12-15 ; Official publication masthead and article byline; 2 Settings; 3 Red-Blue Team Game
- [2] Read the full paper