The field of AI-driven networking and security is rapidly evolving, with a focus on developing innovative solutions to address emerging threats and improve network efficiency. Recent research has explored the use of large language models (LLMs) and AI agents to automate network management, enhance security, and optimize performance. Notably, the development of autonomous networks, powered by AI agents, is gaining traction, with potential applications in 5G and beyond. Furthermore, researchers are investigating the use of decentralized identifiers, secure authenticated key encapsulation protocols, and privacy-preserving authentication mechanisms to enhance security and trust in various networking scenarios. Overall, the field is moving towards a more autonomous, secure, and efficient networking paradigm, driven by advances in AI and machine learning. Noteworthy papers include: A Whole New World, which introduces a novel attack vector exploiting website cloaking techniques to compromise autonomous web-browsing agents. NetGent, which presents an AI-agent framework for automating complex application workflows to generate realistic network traffic datasets. FlexNGIA 2.0, which proposes an Agentic AI-driven Internet architecture that leverages LLM-based AI agents to autonomously orchestrate and evolve the network.
Advances in AI-Driven Networking and Security
Sources
A Whole New World: Creating a Parallel-Poisoned Web Only AI-Agents Can See
NetGent: Agent-Based Automation of Network Application Workflows
Web Fraud Attacks Against LLM-Driven Multi-Agent Systems
Privacy-preserving authentication for military 5G networks
Towards Multi-Platform Mutation Testing of Task-based Chatbots
AmphiKey: A Dual-Mode Secure Authenticated Key Encapsulation Protocol for Smart Grid
A Privacy-Preserving Recommender for Filling Web Forms Using a Local Large Language Model
Throttling Web Agents Using Reasoning Gates
An LLM-enabled semantic-centric framework to consume privacy policies
A Gentle Introduction to Blind signatures: From RSA to Lattice-based Cryptography
AI Agent Communication from Internet Architecture Perspective: Challenges and Opportunities
FlexNGIA 2.0: Redesigning the Internet with Agentic AI - Protocols, Services, and Traffic Engineering Designed, Deployed, and Managed by AI
Situating AI Agents in their World: Aspective Agentic AI for Dynamic Partially Observable Information Systems
Internet 3.0: Architecture for a Web-of-Agents with it's Algorithm for Ranking Agents
AI Agents for Web Testing: A Case Study in the Wild
User Privacy and Large Language Models: An Analysis of Frontier Developers' Privacy Policies
Secure and Trustful Cross-domain Communication with Decentralized Identifiers in 5G and Beyond
Towards Reliable Service Provisioning for Dynamic UAV Clusters in Low-Altitude Economy Networks
20 Years in Life of a Smart Building: A retrospective
Five Blind Men and the Internet: Towards an Understanding of Internet Traffic
Network-level Censorship Attacks in the InterPlanetary File System
Leveraging AI Agents for Autonomous Networks: A Reference Architecture and Empirical Studies
Overcoming DNSSEC Islands of Security: A TLS and IP-Based Certificate Solution