# On the Dual-Use Dilemma in Physical Reasoning and Force

arXiv: https://arxiv.org/abs/2505.18792 · HTML: https://mlsanity.com/paper/2505.18792/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 24 of its 44 references that are in the map.
- First posted on arXiv 24 May 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 24 of 24)

- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Open X-Embodiment: Robotic Learning Datasets and RT-X Models](https://mlsanity.com/paper/2310.08864/) (2023)
- [ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation](https://mlsanity.com/paper/2409.01652/) (2024)
- [RoboPoint: A Vision-Language Model for Spatial Affordance Prediction for Robotics](https://mlsanity.com/paper/2406.10721/) (2024)
- [AutoRT: Embodied Foundation Models for Large Scale Orchestration of Robotic Agents](https://mlsanity.com/paper/2401.12963/) (2024)
- [OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulation](https://mlsanity.com/paper/2505.03912/) (2025)
- [Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics](https://mlsanity.com/paper/2411.13587/) (2024)
- [Rethinking Model Evaluation as Narrowing the Socio-Technical Gap](https://mlsanity.com/paper/2306.03100/) (2023)
- [BadRobot: Jailbreaking Embodied LLM Agents in the Physical World](https://mlsanity.com/paper/2407.20242/) (2024)
- [Safety Guardrails for LLM-Enabled Robots](https://mlsanity.com/paper/2503.07885/) (2025)
- [Exploring the Robustness of Decision-Level Through Adversarial Attacks on LLM-Based Embodied Models](https://mlsanity.com/paper/2405.19802/) (2024)
- [Generating Robot Constitutions & Benchmarks for Semantic Safety](https://mlsanity.com/paper/2503.08663/) (2025)
- [Compromising Embodied Agents with Contextual Backdoor Attacks](https://mlsanity.com/paper/2408.02882/) (2024)
- [On the Vulnerability of LLM/VLM-Controlled Robotics](https://mlsanity.com/paper/2402.10340/) (2024)
- [Robot Collapse: Supply Chain Backdoor Attacks Against VLM-based Robotic Manipulation](https://mlsanity.com/paper/2411.11683/) (2024)
- [Easier Said Than Done: Unpacking Intent-Behavior Gap in Jailbreaking LLM-Based Robots](https://mlsanity.com/paper/2412.16633/) (2024)
- [Towards Forceful Robotic Foundation Models: a Literature Survey](https://mlsanity.com/paper/2504.11827/) (2025)
- [DeliGrasp: Inferring Object Properties with LLMs for Adaptive Grasp Policies](https://mlsanity.com/paper/2403.07832/) (2024)
- [Jailbreaking LLM-Controlled Robots](https://mlsanity.com/paper/2410.13691/) (2024)
- ["Can you be my mum?": Manipulating Social Robots in the Large Language Models Era](https://mlsanity.com/paper/2501.04633/) (2025)
- [The State of Robot Motion Generation](https://mlsanity.com/paper/2410.12172/) (2024)
- [Unfettered Forceful Skill Acquisition with Physical Reasoning and Coordinate Frame Labeling](https://mlsanity.com/paper/2505.09731/) (2025)

## Abstract (arXiv)

Humans learn how and when to apply forces in the world via a complex physiological and psychological learning process. Attempting to replicate this in vision-language models (VLMs) presents two challenges: VLMs can produce harmful behavior, which is particularly dangerous for VLM-controlled robots which interact with the world, but imposing behavioral safeguards can limit their functional and ethical extents. We conduct two case studies on safeguarding VLMs which generate forceful robotic motion, finding that safeguards reduce both harmful and helpful behavior involving contact-rich manipulation of human body parts. Then, we discuss the key implication of this result--that value alignment may impede desirable robot capabilities--for model evaluation and robot learning.