Ecology of AI Agents: Collaboration Creates a Population Threshold for Takeoff
cs.AI, cond-mat.dis-nn, cs.MA, physics.bio-ph
Submitted: 2026-10-08
Updated: 2026-10-08
Comments: 22 pages, 11 figures, 1 table
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Measuring AI Agents' Progress on Multi-Step Cyber Attack Scenarios
- Multi-Agent Risks from Advanced AI
- When Is Collective Intelligence a Lottery? Multi-Agent Scaling Laws for Memetic Drift in LLMs
- Flag Game: A Toy Model for Mechanistic Swarm Interpretability
- Evaluating Language-Model Agents on Realistic Autonomous Tasks
- Dive into the Agent Matrix: A Realistic Evaluation of Self-Replication Risk in LLM Agents
- RepliBench: Evaluating the Autonomous Replication Capabilities of Language Model Agents
- Language Models Can Autonomously Hack and Self-Replicate
- Frontier Models are Capable of In-context Scheming
- Peer-Preservation in Frontier Models
- More Agents Is All You Need
- Scaling Large Language Model-based Multi-Agent Collaboration
- Two Heads are Better Than One: Test-time Scaling of Multi-agent Collaborative Reasoning
- Towards a Science of Scaling Agent Systems
- Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
- SEC-bench Pro: Can Language Models Solve Long-Horizon Software Security Tasks?
- BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems