LLM-Based Role Inference for Scalable Multi-Domain Access Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional RBAC systems require significant manual oversight for user role management, leading to inefficiencies as the number of users and roles scales, and changes in user status necessitate manual role modifications, which is taxing and error-prone.

Innovation Solution

Integrate a large language model (LLM) to dynamically interpret and analyze relationships to infer and implement RBAC, enabling real-time access management and role assignments based on user interactions and relationships.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional RBAC systems are used, then access control can be managed based on predefined roles, but manual role management becomes burdensome and error-prone as the number of users and roles scales

Engineering Contradiction:

Improveaccess control adaptabilityVSAvoidrole management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system enables self-service role assignment by having users provide descriptions of their needed access, and the LLM automatically infers appropriate roles and assigns them without manual intervention from administrators. This transforms the manual role management process into an automated self-service system.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual process of role assignment and permission management with an automated LLM-based system that uses natural language processing to infer roles and assign permissions, eliminating the need for manual administrative overhead.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If manual role assignments are performed, then access control can be precisely controlled, but the process is time-consuming and prone to errors

Engineering Contradiction:

Improveaccess control accuracyVSAvoidrole management time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

Users independently initiate the role assignment process by describing their access needs, and the system automatically completes the role inference and assignment without requiring administrator time or manual intervention, thereby reducing time loss while maintaining accuracy.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system provides feedback by presenting users with inferred role options based on their descriptions and allowing them to confirm or modify the assignment, ensuring accuracy while reducing the time required compared to manual verification processes.

Inventive Principle:

Principle #23Feedback

3Ease of operation

If users are not assigned necessary roles, then access control can be simplified, but users cannot access their necessary organizational resources

Engineering Contradiction:

Improveaccess control easeVSAvoidresource access efficiency

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

Users can independently describe their needed access without requiring manual role assignment from administrators. The LLM automatically infers the appropriate roles and assigns permissions, enabling users to access resources efficiently while maintaining simplified control mechanisms.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically assigns roles based on real-time user needs expressed in natural language, rather than relying on static pre-defined roles, allowing flexible and adaptive access control that responds to changing user requirements.

Inventive Principle:

Principle #15Dynamics

4Adaptability or versatility

If users are over-assigned roles, then access control can be more permissive, but users gain access to resources they should not be permitted to access

Engineering Contradiction:

Improveaccess flexibilityVSAvoidunauthorized access risk

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The system uses feedback mechanisms where users review and confirm the inferred roles before final assignment, allowing them to correct any over-assignments or inaccuracies, thereby preventing unauthorized access while maintaining flexibility.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs partial role assignment by inferring only the necessary roles based on user descriptions rather than assigning all possible roles, and includes a verification step that allows users to reject inappropriate assignments, preventing excessive access while maintaining flexibility.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20250373611A1Multi-domain, role-based access control using large language models

Publication Date: 2025.12.04 AMERICAN EXPRESS TRAVEL RELATED SERVICES CO INC

AI summary

Disclosed are various embodiments for multi-domain, role-based access control (RBAC) using large language models. Various embodiments can receive an access request from a client device associated with a user. Various embodiments can then send a prompt to identify a user role for the user to an agent of a machine learning model. Various embodiments can receive the user role for the user from the agent. The various embodiments can determine the capability for the user to access a resource by comparing the user role for the user to allowed user roles for the resource. Various embodiments can then send an access response to the client device that indicates whether the user can access the resource.