# Circumventing interpretability: How to defeat mind-readers

arXiv: https://arxiv.org/abs/2212.11415 · HTML: https://mlsanity.com/paper/2212.11415/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 53 references that are in the map.
- First posted on arXiv 21 Dec 2022.

## Builds on (most cited first, 12 of 12)

- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [Universal Transformers](https://mlsanity.com/paper/1807.03819/) (2018)
- [Weight Poisoning Attacks on Pre-trained Models](https://mlsanity.com/paper/2004.06660/) (2020)
- [Building machines that adapt and compute like brains](https://mlsanity.com/paper/1711.04203/) (2017)
- [An Interpretability Illusion for BERT](https://mlsanity.com/paper/2104.07143/) (2021)
- [Planting Undetectable Backdoors in Machine Learning Models](https://mlsanity.com/paper/2204.06974/) (2022)
- [An overview of 11 proposals for building safe advanced AI](https://mlsanity.com/paper/2012.07532/) (2020)
- [Interval timing in deep reinforcement learning agents](https://mlsanity.com/paper/1905.13469/) (2019)
- [Adversarially trained neural representations may already be as robust as corresponding biological neural representations](https://mlsanity.com/paper/2206.11228/) (2022)
- [Trojan Horse Training for Breaking Defenses against Backdoor Attacks in Deep Learning](https://mlsanity.com/paper/2203.15506/) (2022)

## Built on by (most cited first, 6 of 6)

- [Mechanistic Interpretability for AI Safety -- A Review](https://mlsanity.com/paper/2404.14082/) (2024)
- [Don't trust your eyes: on the (un)reliability of feature visualizations](https://mlsanity.com/paper/2306.04719/) (2023)
- [Concept-Level Explainability for Auditing & Steering LLM Responses](https://mlsanity.com/paper/2505.07610/) (2025)
- [AI Testing Should Account for Sophisticated Strategic Behaviour](https://mlsanity.com/paper/2508.14927/) (2025)
- [Unraveling Token Prediction Refinement and Identifying Essential Layers in Language Models](https://mlsanity.com/paper/2501.15054/) (2025)
- [Correcting Influence: Unboxing LLM Outputs with Orthogonal Latent Spaces](https://mlsanity.com/paper/2605.12809/) (2026)

## Abstract (arXiv)

The increasing capabilities of artificial intelligence (AI) systems make it ever more important that we interpret their internals to ensure that their intentions are aligned with human values. Yet there is reason to believe that misaligned artificial intelligence will have a convergent instrumental incentive to make its thoughts difficult for us to interpret. In this article, I discuss many ways that a capable AI might circumvent scalable interpretability methods and suggest a framework for thinking about these potential future risks.