Closing the Responsibility Gap in AI-based Network Management: An Intelligent Audit System Approach

TL;DR

This work tackles the accountability gap in AI-based network management by proposing a dual-model audit framework that combines Deep Reinforcement Learning (DRL) for identifying AI agents and a machine learning (ML) component for inferring network conditions. It formalizes responsibility through a dynamic MDP-based modeling and an iterative model-mapping process to generate explainable, time-stamped audit trails, all implemented within a Zero Trust Data Plane that isolates vendor automation tools. The framework is validated in simulation, achieving 96% DRL agent-identification accuracy and 83% ML condition-learning accuracy, demonstrating the viability of auditable AI-driven network management for future 5G/6G deployments. Overall, the approach advances practical governance for AI in networking by enabling traceable decision reasoning, vendor accountability, and safer, more transparent autonomous network operations.

Abstract

Existing network paradigms have achieved lower downtime as well as a higher Quality of Experience (QoE) through the use of Artificial Intelligence (AI)-based network management tools. These AI management systems, allow for automatic responses to changes in network conditions, lowering operation costs for operators, and improving overall performance. While adopting AI-based management tools enhance the overall network performance, it also introduce challenges such as removing human supervision, privacy violations, algorithmic bias, and model inaccuracies. Furthermore, AI-based agents that fail to address these challenges should be culpable themselves rather than the network as a whole. To address this accountability gap, a framework consisting of a Deep Reinforcement Learning (DRL) model and a Machine Learning (ML) model is proposed to identify and assign numerical values of responsibility to the AI-based management agents involved in any decision-making regarding the network conditions, which eventually affects the end-user. A simulation environment was created for the framework to be trained using simulated network operation parameters. The DRL model had a 96% accuracy during testing for identifying the AI-based management agents, while the ML model using gradient descent learned the network conditions at an 83% accuracy during testing.

Figures (3)

• Figure 1: Decomposition of Model to MDP
• Figure 2: Proposed framework for assigning responsibility to the AI gap
• Figure 3: Proposed framework for Model Mapping