Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture
Halil Burak Noyan
cs.AI, cs.CR
Submitted: 2026-07-24
Comments: Published at the Second Workshop on Agents in the Wild: Safety, Security, and Beyond (AIWILD) at ICML 2026
Code: https://github.com/UKGovernmentBEIS/control-arena
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Alignment faking in large language models
- Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
- Toolformer: Language Models Can Teach Themselves to Use Tools
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- Frontier Models are Capable of In-context Scheming
- A Systematic Security Evaluation of OpenClaw and Its Variants
- ReAct: Synergizing Reasoning and Acting in Language Models
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
- 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
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection