# Corrigibility with Utility Preservation

arXiv: https://arxiv.org/abs/1908.01695 · HTML: https://mlsanity.com/paper/1908.01695/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Aug 2019.

## Built on by (most cited first, 8 of 8)

- [Beyond Preferences in AI Alignment](https://mlsanity.com/paper/2408.16984/) (2024)
- [Human Control: Definitions and Algorithms](https://mlsanity.com/paper/2305.19861/) (2023)
- [Emergent Risk Awareness in Rational Agents under Resource Constraints](https://mlsanity.com/paper/2505.23436/) (2025)
- [Counterfactual Planning in AGI Systems](https://mlsanity.com/paper/2102.00834/) (2021)
- [Towards Shutdownable Agents via Stochastic Choice](https://mlsanity.com/paper/2407.00805/) (2024)
- [Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs](https://mlsanity.com/paper/2604.17502/) (2026)
- [Password-Activated Shutdown Protocols for Misaligned Frontier Agents](https://mlsanity.com/paper/2512.03089/) (2025)
- [Reframing AI Loss of Control: What Control Is, How to Have It, How to Lose It](https://mlsanity.com/paper/2606.12442/) (2026)

## Abstract (arXiv)

Corrigibility is a safety property for artificially intelligent agents. A corrigible agent will not resist attempts by authorized parties to alter the goals and constraints that were encoded in the agent when it was first started. This paper shows how to construct a safety layer that adds corrigibility to arbitrarily advanced utility maximizing agents, including possible future agents with Artificial General Intelligence (AGI). The layer counter-acts the emergent incentive of advanced agents to resist such alteration. A detailed model for agents which can reason about preserving their utility function is developed, and used to prove that the corrigibility layer works as intended in a large set of non-hostile universes. The corrigible agents have an emergent incentive to protect key elements of their corrigibility layer. However, hostile universes may contain forces strong enough to break safety features. Some open problems related to graceful degradation when an agent is successfully attacked are identified. The results in this paper were obtained by concurrently developing an AGI agent simulator, an agent model, and proofs. The simulator is available under an open source license. The paper contains simulation results which illustrate the safety related properties of corrigible AGI agents in detail.