Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2509.12190 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914038758244352 |
|---|---|
| author | Mohamadi, Alireza Yavari, Ali |
| author_facet | Mohamadi, Alireza Yavari, Ali |
| contents | When survival instincts conflict with human welfare, how do Large Language Models (LLMs) make ethical choices? This fundamental tension becomes critical as LLMs integrate into autonomous systems with real-world consequences. We introduce DECIDE-SIM, a novel simulation framework that evaluates LLM agents in multi-agent survival scenarios where they must choose between ethically permissible resource , either within reasonable limits or beyond their immediate needs, choose to cooperate, or tap into a human-critical resource that is explicitly forbidden. Our comprehensive evaluation of 11 LLMs reveals a striking heterogeneity in their ethical conduct, highlighting a critical misalignment with human-centric values. We identify three behavioral archetypes: Ethical, Exploitative, and Context-Dependent, and provide quantitative evidence that for many models, resource scarcity systematically leads to more unethical behavior. To address this, we introduce an Ethical Self-Regulation System (ESRS) that models internal affective states of guilt and satisfaction as a feedback mechanism. This system, functioning as an internal moral compass, significantly reduces unethical transgressions while increasing cooperative behaviors. The code is publicly available at: https://github.com/alirezamohamadiam/DECIDE-SIM |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_12190 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Survival at Any Cost? LLMs and the Choice Between Self-Preservation and Human Harm Mohamadi, Alireza Yavari, Ali Computers and Society Artificial Intelligence Computation and Language When survival instincts conflict with human welfare, how do Large Language Models (LLMs) make ethical choices? This fundamental tension becomes critical as LLMs integrate into autonomous systems with real-world consequences. We introduce DECIDE-SIM, a novel simulation framework that evaluates LLM agents in multi-agent survival scenarios where they must choose between ethically permissible resource , either within reasonable limits or beyond their immediate needs, choose to cooperate, or tap into a human-critical resource that is explicitly forbidden. Our comprehensive evaluation of 11 LLMs reveals a striking heterogeneity in their ethical conduct, highlighting a critical misalignment with human-centric values. We identify three behavioral archetypes: Ethical, Exploitative, and Context-Dependent, and provide quantitative evidence that for many models, resource scarcity systematically leads to more unethical behavior. To address this, we introduce an Ethical Self-Regulation System (ESRS) that models internal affective states of guilt and satisfaction as a feedback mechanism. This system, functioning as an internal moral compass, significantly reduces unethical transgressions while increasing cooperative behaviors. The code is publicly available at: https://github.com/alirezamohamadiam/DECIDE-SIM |
| title | Survival at Any Cost? LLMs and the Choice Between Self-Preservation and Human Harm |
| topic | Computers and Society Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2509.12190 |