LLMAC: A Global and Explainable Access Control Framework with Large Language Model

2026-01-09

Consumer Communications and Networking Conference

AI Summary

This work proposes the first unified access control framework grounded in large language models to address the limitations of traditional models—such as RBAC, ABAC, and DAC—in handling dynamic, context-sensitive security policies in modern systems. By fine-tuning Mistral-7B with a synthetically generated dataset, the framework integrates multiple classical paradigms to achieve comprehensive understanding and reasoning over complex policies, including ownership verification, version control, workflow constraints, and dynamic separation of duties. The approach attains a decision accuracy of 98.5% while maintaining practical response latency and computational overhead, significantly outperforming conventional methods. Furthermore, it provides human-readable justifications for its decisions, thereby achieving both high accuracy and strong interpretability.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Computer Vision: Large Vision Models

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Security and Privacy: Authentication, authorization, and access control

Abstract

Today’s business organizations need access control systems that can handle complex, changing security requirements that go beyond what traditional methods can manage. Current approaches, such as Role-Based Access Control (RBAC), Attribute-Based Access Control (ABAC), and Discretionary Access Control (DAC), were designed for specific purposes. They cannot effectively manage the dynamic, situation-dependent workflows that modern systems require. In this research, we introduce LLMAC, a new unified approach using Large Language Models (LLMs) to combine these different access control methods into one comprehensive, understandable system. We used an extensive synthetic dataset that represents complex real-world scenarios, including policies for ownership verification, version management, workflow processes, and dynamic role separation. Using Mistral 7B, our trained LLM model achieved outstanding results with 98.5% accuracy, significantly outperforming traditional methods (RBAC: 14.5%, ABAC: 58.5%, DAC: 27.5%) while providing clear, human readable explanations for each decision. Performance testing shows that the system can be practically deployed with reasonable response times and computing resources.

Problem

• Access Control
• Dynamic Workflows
• Explainability
• Security Policies
• Role-Based Access Control

Innovation

• Large Language Model
• Access Control
• Explainability
• Unified Framework
• Dynamic Policy