# Reliability Theory for AI Control

arXiv: https://arxiv.org/abs/2609.26419 · HTML: https://mlsanity.com/paper/2609.26419/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 30 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [Ctrl-Z: Controlling AI Agents via Resampling](https://mlsanity.com/paper/2504.10374/) (2025)
- [Evaluating Control Protocols for Untrusted AI Agents](https://mlsanity.com/paper/2511.02997/) (2025)
- [Ensemble Monitoring for AI Control: Diverse Signals Outweigh More Compute](https://mlsanity.com/paper/2605.15377/) (2026)
- [Assurance of AI Systems From a Dependability Perspective](https://mlsanity.com/paper/2407.13948/) (2024)
- [Factor(T,U): Factored Cognition Strengthens Monitoring of Untrusted AI](https://mlsanity.com/paper/2512.02157/) (2025)
- [Adapting Probabilistic Risk Assessment for AI](https://mlsanity.com/paper/2504.18536/) (2025)
- [N-Version Programming with Coding Agents](https://mlsanity.com/paper/2606.20158/) (2026)
- [GDM AI Control Roadmap](https://mlsanity.com/paper/2607.13087/) (2026)
- [Retrying vs Resampling in AI Control](https://mlsanity.com/paper/2605.26047/) (2026)
- [Attack Selection in Agentic AI Control Evaluations Meaningfully Decreases Safety](https://mlsanity.com/paper/2606.06529/) (2026)

## Built on by (most cited first, 1 of 1)

- [A Competing-Hazards Systematization of Loss of Control in Autonomous Agents](https://mlsanity.com/paper/2609.38411/) (2026)

## Abstract (arXiv)

Reliability theory gives a mature language for layered systems, but its formal tools are not yet standard in frontier AI control. We apply them to Google DeepMind's defenses against rogue deployment. The same control stack can have cubic, quadratic, or linear rare-failure suppression depending on its failure domains. Birnbaum importance identifies which component improvements buy the most nominal reliability, while prevention changes the population on which recovery is demanded. These results give concrete guidance about what to separate, improve, measure, and test.