[2302.10329] Harms from Increasingly Agentic Algorithmic Systems

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2302.10329] Harms from Increasingly Agentic Algorithmic Systems Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Computers and Society arXiv:2302.10329 (cs) [Submitted on 20 Feb 2023 ( v1 ), last revised 12 May 2023 (this version, v2)] Title: Harms from Increasingly Agentic Algorithmic Systems Authors: Alan Chan , Rebecca Salganik , Alva Markelius , Chris Pang , Nitarshan Rajkumar , Dmitrii Krasheninnikov , Lauro Langosco , Zhonghao He , Yawen Duan , Micah Carroll , Michelle Lin ,

related reading

• [2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingarxiv.org
• [2502.14143] Multi-Agent Risks from Advanced AIarxiv.org
• Agents of Chaosarxiv.org
• [2304.03279] Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmarkarxiv.org
• [2507.13616] From Firms to Computation: AI Governance and the Evolution of Institutionsarxiv.org
• [2312.15159] Understanding the Potential of FPGA-Based Spatial Acceleration for Large Language Model Inferencearxiv.org
• [2303.13360] Towards the Scalable Evaluation of Cooperativeness in Language Modelsarxiv.org
• [2305.18654] Faith and Fate: Limits of Transformers on Compositionalityarxiv.org
• [2402.04249] HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalarxiv.org
• [2406.04219] Multi-Agent Imitation Learning: Value is Easy, Regret is Hardarxiv.org
• [2512.18552] Toward Training Superintelligent Software Agents through Self-Play SWE-RLarxiv.org
• [2309.01809] Are Emergent Abilities in Large Language Models just In-Context Learning?arxiv.org