Alignment of Complex Systems Research Group – ACS Research Program
Our research investigates the dynamics of intelligence across multiple scales—from individual agents to complex, civilization-scale systems. We aim to develop the foundational understanding and conceptual frameworks necessary to navigate the systemic risks posed by advanced AI and ensure that the evolution of human-AI ecosystems remains aligned with human values. Our research seeks to build towards the following goals: A primary focus of our research addresses the systemic risks posed by the incremental advancement and proliferation of AI. While much AI safety discourse focuses on abrupt takeover scenarios, we investigate the dynamics of 'gradual disempowerment.' This describes a process where human influence over crucial societal systems—including the economy, culture, and governance—is irreversibly eroded, even without coordinated power-seeking by AI systems. Our work aims to develop methods for measuring disempowerment and exploring technical and governance strategies to maintain me
Alignment of Complex Systems Research Group – ACS Research Program ACS Research Program Our research investigates the dynamics of intelligence across multiple scales — from individual agents to complex, civilization-scale systems. We aim to develop the foundational understanding and conceptual frameworks necessary to navigate the systemic risks posed by advanced AI and ensure that the evolution of human-AI ecosystems remains aligned with human values. Our two main research areas are: Gradual Disempowerment — understanding and mitigating the systemic existential risks from incremental AI advanc
saved by
related reading
- Gradual Disempowermentgradual-disempowerment.ai
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Sequent: scale and automation for higher confidence in alignment — AI Alignment Forumalignmentforum.org
- Center for the Alignment of AI Alignment Centersalignmentalignment.ai
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- Gradual Disempowerment: Systemic Existential Risks from Incremental AI Developmentarxiv.org
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Automated Alignment Researchers: Using large language models to scale scalable oversight \ Anthropicanthropic.com
- The Artificial Selftheartificialself.ai
- Focus areas for The Anthropic Instituteanthropic.com
- [2502.14143] Multi-Agent Risks from Advanced AIarxiv.org
- Research Areas in AI Control (The Alignment Project by UK AISI) — AI Alignment Forumalignmentforum.org