# Categorizing Wireheading in Partially Embedded Agents

arXiv: https://arxiv.org/abs/1906.09136 · HTML: https://mlsanity.com/paper/1906.09136/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jun 2019.

## Built on by (most cited first, 5 of 5)

- [Counterfactual Planning in AGI Systems](https://mlsanity.com/paper/2102.00834/) (2021)
- [A Model-Free Universal AI](https://mlsanity.com/paper/2602.23242/) (2026)
- [Does Self-Evaluation Enable Wireheading in Language Models?](https://mlsanity.com/paper/2511.23092/) (2025)
- [The Tao of Agency: Autotelic AI, Embedded Agency and Dissolution of the Self](https://mlsanity.com/paper/2606.19924/) (2026)
- [Modification-Considering Value Learning for Reward Hacking Mitigation in RL](https://mlsanity.com/paper/2606.28955/) (2026)

## Abstract (arXiv)

textitEmbedded agents are not explicitly separated from their environment, lacking clear I/O channels. Such agents can reason about and modify their internal parts, which they are incentivized to shortcut or textitwirehead in order to achieve the maximal reward. In this paper, we provide a taxonomy of ways by which wireheading can occur, followed by a definition of wirehead-vulnerable agents. Starting from the fully dualistic universal agent AIXI, we introduce a spectrum of partially embedded agents and identify wireheading opportunities that such agents can exploit, experimentally demonstrating the results with the GRL simulation platform AIXIjs. We contextualize wireheading in the broader class of all misalignment problems - where the goals of the agent conflict with the goals of the human designer - and conjecture that the only other possible type of misalignment is specification gaming. Motivated by this taxonomy, we define wirehead-vulnerable agents as embedded agents that choose to behave differently from fully dualistic agents lacking access to their internal parts.