# Some[Body] Must Receive That Pain for Agent Accountability

arXiv: https://arxiv.org/abs/2605.16872 · HTML: https://mlsanity.com/paper/2605.16872/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 70 references that are in the map.
- First posted on arXiv 16 May 2026.

## Builds on (most cited first, 26 of 26)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning](https://mlsanity.com/paper/2308.08747/) (2023)
- [Stealing Machine Learning Models via Prediction APIs](https://mlsanity.com/paper/1609.02943/) (2016)
- [Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training](https://mlsanity.com/paper/2401.05566/) (2024)
- [Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks](https://mlsanity.com/paper/2404.02151/) (2024)
- [Frontier Models are Capable of In-context Scheming](https://mlsanity.com/paper/2412.04984/) (2024)
- [The Forward-Forward Algorithm: Some Preliminary Investigations](https://mlsanity.com/paper/2212.13345/) (2022)
- [Agentic Misalignment: How LLMs Could Be Insider Threats](https://mlsanity.com/paper/2510.05179/) (2025)
- [Consciousness in Artificial Intelligence: Insights from the Science of Consciousness](https://mlsanity.com/paper/2308.08708/) (2023)
- [I Know What You Trained Last Summer: A Survey on Stealing Machine Learning Models and Defences](https://mlsanity.com/paper/2206.08451/) (2022)
- [Emotion Concepts and their Function in a Large Language Model](https://mlsanity.com/paper/2604.07729/) (2026)
- [Understanding accountability in algorithmic supply chains](https://mlsanity.com/paper/2304.14749/) (2023)
- [Taking AI Welfare Seriously](https://mlsanity.com/paper/2411.00986/) (2024)
- [Incomplete Tasks Induce Shutdown Resistance in Some Frontier LLMs](https://mlsanity.com/paper/2509.14260/) (2025)
- [Frontier AI systems have surpassed the self-replicating red line](https://mlsanity.com/paper/2412.12140/) (2024)
- [Mortal Computation: A Foundation for Biomimetic Intelligence](https://mlsanity.com/paper/2311.09589/) (2023)
- [Sovereign Agents: Towards Infrastructural Sovereignty and Diffused Accountability in Decentralized AI](https://mlsanity.com/paper/2602.14951/) (2026)
- [How to Count AIs: Individuation and Liability for AI Agents](https://mlsanity.com/paper/2603.10028/) (2026)
- [AI Wellbeing](https://mlsanity.com/paper/2509.11913/) (2025)
- [Do Artificial Reinforcement-Learning Agents Matter Morally?](https://mlsanity.com/paper/1410.8233/) (2014)

## Abstract (arXiv)

AI agents increasingly act consequentially in the real world. This creates a problem we call consequence reception: harm occurs, the producing system is identified, yet no continuing agent receives consequences in a way that changes future behavior. Pain, understood mechanistically as a corrective feedback signal, is foundational to canonical theories of punishment -- deterrence, rehabilitation, retribution, and incapacitation all assume a continuing locus that registers the signal and updates behavior. That, in turn, requires a body for the signal to land on: a boundary whose integrity it protects, a locus where it accumulates, consolidation that converts episodic signal into durable update, and a substrate that responds by altering future action. Current LLM agents -- software-defined composites of weights, prompts, tools, memory, and credentials, freely swapped, copied, reset, and reassembled -- satisfy none of these conditions. The two prevailing legal responses therefore fail to achieve consequence reception. The thin-identity agent-principal dyad has a body but no consequence--agency coupling: the human bears pain for behaviors beyond their control -- Elish's moral crumple zone. The thick-identity Arbel et al.'s Algorithmic Corporation creates legally legible entities but does not guarantee that any AI decision architecture receives pain as a behavioral signal. Achieving consequence-agency coupling is therefore a sociotechnical infrastructural problem, not only a legal one. Until such architectures exist, high-stakes AI deployment should remain tethered to accountable human principals with meaningful control, proportional liability, and authority to constrain or terminate the agent. If some body does not receive the pain by design, some body will receive it by default.